Metonymy and pragmatic
inferencing in mental health
Andrew Moors
Doctor of Philosophy
2015
CERTIFICATE OF ORIGINAL AUTHORSHIP
I certify that the work in this thesis has not previously been submitted for a degree nor
has it been submitted as part of requirements for a degree except as fully
acknowledged within the text.
I also certify that the thesis has been written by me. Any help that I have received in my
research work and the preparation of the thesis itself has been acknowledged. In
addition, I certify that all information sources and literature used are indicated in the
thesis.
Signature of Student:
Date
Front pages
i
Front pages
ii
ACKNOWLEDGEMENTS
I thank the participants, for their courage and their curiosity, for sharing their stories
and making the study possible. I thank Professor Jane Stein-Parbury, for her generous
and astute supervision throughout the last eight years. I also thank Dr Michael Roche,
Dr Michael Carey and Professor Diana Slade for their contributions. I thank my former
clinical colleagues, the Schizophrenia Fellowship of NSW and the NSW Consumer
Advisory Group for their help in recruiting participants. I thank my fellow postgraduate
students for their support and comments. I thank Dr Suellen Allen, for her contributions
to my academic and professional life. I acknowledge the work of the late Michael White
in the formation of many of the ideas that have guided my practice and this research. I
thank my parents, Fran and Ted Moors. I thank Dr Jean Ellard, for her robust intellect,
and her loving friendship over 27 years. I thank my children, Eitan and Ella, for the joy
they bring to every day, and for letting Adi go to uni so much. And I thank my partner,
Dr David Mor, for the practical and emotional support which helped me undertake this
study, and for his love.
This thesis is dedicated to people who live with mental illness, and the memory of
those who went before, with respect for their courage, humour and grace.
Front pages
iii
Front pages
iv
Table of Contents
ACKNOWLEDGEMENTS ............................................................................................. iii
ABSTRACT ................................................................................................................. xi
CHAPTER 1: INTRODUCTION .................................................................................... 1
Key terms ................................................................................................................. 2
Aim and Objectives of the study ................................................................................ 2
Research questions .................................................................................................. 2
Background............................................................................................................... 3
Language and mental illness................................................................................. 3
Contemporary approaches to language ................................................................ 4
Cognitive linguistics ........................................................................................... 4
Pragmatics ........................................................................................................ 6
Metonymy ............................................................................................................. 7
Metonymy and mental illness ............................................................................ 8
Metonymy and therapeutic approaches ............................................................. 9
The experience of metonymic language for people with mental illness ............ 10
Study overview ....................................................................................................... 11
CHAPTER 2: LANGUAGE AND MENTAL ILLNESS ................................................... 13
Introduction ............................................................................................................. 13
Diagnosis ................................................................................................................ 13
Speech production: observation and inference.................................................... 13
Speech production: content and form .................................................................. 16
Speech comprehension ...................................................................................... 18
Diagnostic heterogeneity ..................................................................................... 19
Figurative language............................................................................................. 19
Delivery of treatment ............................................................................................... 20
Evaluation of treatment ........................................................................................... 22
Medico-legal implications ........................................................................................ 23
Front pages
v
Historical approaches to language and mental illness ............................................. 25
Early twentieth century approaches .................................................................... 25
1930s and 1940s: Metonymic distortion .............................................................. 26
Metonymic speech in subsequent psychiatric discourse .................................. 28
Studies from the second half of the twentieth century ......................................... 29
Approaches to language and mental health beyond psychiatry ............................... 32
Nursing ............................................................................................................... 32
Conversational therapy and systemic functional linguistics ................................. 33
Psycholinguistics and Neurolinguistics ................................................................ 34
Narrative therapy ................................................................................................ 38
Language and identity as experienced by people who live with mental illness ........ 40
Stigma................................................................................................................. 42
Conclusion .............................................................................................................. 43
CHAPTER 3: METONYMY ......................................................................................... 45
Earlier definitions .................................................................................................... 45
Jakobson ............................................................................................................ 46
Cognitive linguistic definitions ................................................................................. 47
A working definition ............................................................................................. 50
Target and source ........................................................................................... 51
Mapping .......................................................................................................... 51
Asymmetry ...................................................................................................... 52
Functional Domain .......................................................................................... 52
Pragmatic function .......................................................................................... 53
Activation ........................................................................................................ 53
Types of metonymy................................................................................................. 54
Referential metonymy ......................................................................................... 54
Predicational metonymy ...................................................................................... 54
Illocutionary metonymy ....................................................................................... 55
Uses of metonymy in nursing contexts .................................................................... 57
Front pages
vi
Metonymy and mental illness .................................................................................. 61
Labelling theory and stigma ................................................................................ 62
Metonymy and narrative therapy ......................................................................... 65
Conclusion .............................................................................................................. 66
CHAPTER 4: METHOD .............................................................................................. 69
Introduction ............................................................................................................. 69
Participant selection and recruitment ...................................................................... 70
Data Collection ....................................................................................................... 71
Biographical Narrative Interpretive Method ......................................................... 72
Interviews ............................................................................................................ 73
Initial interviews ............................................................................................... 73
Second interviews ........................................................................................... 74
Data Management .................................................................................................. 75
Data Analysis .......................................................................................................... 76
Template analysis ............................................................................................... 77
Templates ........................................................................................................... 77
Template 1: Themes ....................................................................................... 78
Template 2: Cognitive linguistic terms ............................................................. 79
Template 3: Andreasen’s Thought, Language and Communication Scale ....... 81
Template 4: Narrative Modes: Wengraf’s ‘Textsort’ categories ........................ 83
Synthesis of Template Analyses ......................................................................... 83
Ethical considerations ............................................................................................. 84
Informed and voluntary consent .......................................................................... 84
Risk and benefit .................................................................................................. 85
Participant distress .............................................................................................. 85
Confidentiality ..................................................................................................... 86
Reflexivity ............................................................................................................... 87
Conclusion .............................................................................................................. 89
CHAPTER 5: RESULTS ............................................................................................. 91
Front pages
vii
Participants ............................................................................................................. 92
Formal mental health diagnoses ......................................................................... 93
Treatment ........................................................................................................... 93
Interviews ............................................................................................................... 94
Themes................................................................................................................... 95
Identity ................................................................................................................ 95
Stigma............................................................................................................... 100
Experience of symptoms ................................................................................... 104
Experience of treatment .................................................................................... 110
Positive experiences ..................................................................................... 110
Negative experiences .................................................................................... 112
Insight ............................................................................................................... 121
Recovery........................................................................................................... 124
Experience of language ........................................................................................ 129
Metonymy: Quick quiz ....................................................................................... 129
Metonymy, metaphor and other language phenomena ..................................... 132
Narrative ........................................................................................................... 136
Participants’ own observations of their use of language .................................... 138
Conclusion ............................................................................................................ 141
Metonymic influences on participants’ experience ............................................. 141
Metonymic aspects of participants’ language .................................................... 143
CHAPTER 6: DISCUSSION ..................................................................................... 145
The influence of metonymic language on the experience of people who live with
mental illness ........................................................................................................ 146
Identity .............................................................................................................. 146
Visibility ............................................................................................................. 149
Stigma............................................................................................................... 152
Self-stigma .................................................................................................... 153
Stigma from health workers ........................................................................... 154
Front pages
viii
Other metonymic influences on the experience of treatment ............................. 156
Metonymy and the experience of symptoms ..................................................... 159
The influence of metonymy on the speech production and speech comprehension of
people who live with mental illness ....................................................................... 161
Metonymic speech ............................................................................................ 162
Comprehension of metonymy ........................................................................... 162
Linguistic and conceptual metonymy in participants’ ‘everyday’ language ......... 164
Narrative ........................................................................................................... 165
Contribution to cognitive linguistics ....................................................................... 167
Limitations of the study ......................................................................................... 168
Implications for further research ............................................................................ 169
Implications for practice ........................................................................................ 170
Conclusion ............................................................................................................ 172
APPENDICES .......................................................................................................... 175
REFERENCES ......................................................................................................... 181
List of tables
Table 4.1 Themes
76
Table 4.2 Cognitive linguistic terms
77
Table 4.3 Andreasen’s TLC Scale
79
Table 4.4 Wengraf’s Textsort categories
80
Table 5.1 Participant characteristics
88
Table 5.2 Interview data
90
Front pages
ix
Front pages
x
ABSTRACT
Language plays a key role in the assessment, diagnosis and treatment of mental
illness. Pragmatically, how language functions in mental health care is heavily
dependent on inferences that are drawn, both by the people experiencing mental
illness, and by the clinicians assisting them. Metonymy is a language practice where
one term, the source, is used to stand for another, the target, which is closely related to
it conceptually. The use of a diagnostic category to refer to a person, for example, a
schizophrenic, is a metonymic transfer of meaning. Metonymy occurs in everyday
language and in clinical settings, and contributes to how inferences are drawn. It
functions because it draws attention to a specific feature with minimal cognitive
expenditure. Metonymy contributes to effective communication in contexts where the
language-using group share understanding of what is deemed to be salient, and where
economy of speech is valued. Nursing provides such a context. Though the intended
meaning of a metonymic reference may be understood in any communication, there is
the potential for attributes of the source term to carry over into the interpretation of the
target meaning and lead to unintended inferences. In the mental health context,
metonymy can be associated with negative outcomes in relation to labelling and the
experience of stigma. Strategies have been implemented to modify language to reduce
the influence of this labelling, but to date, they have not made use of the insights of
cognitive linguistics.
This study is an investigation of the influence of metonymic language on the
experience of people who live with mental illness. The study draws on nineteen
interviews with ten participants with lived experience of psychosis. Interviews were
conducted using a biographical narrative method and analysed thematically, to elicit
participants’ experiences of language. Their language was also analysed for cognitive
and clinical linguistic features. Participants reported experiencing language used about
them in ways that foregrounded their diagnoses, including stigmatising language. They
also reported experiences when they felt that legitimate concerns or requests they
expressed were disregarded because of the foregrounding of their diagnoses.
Participants described the beneficial effect they experienced when mental health
workers used language that maintained the person’s identity separate from the illness.
Metonymy and pragmatic inferencing underpin the experiences of language reported
by people who live with mental illness, with both positive and negative effects. Closer
attention to the influence of language can contribute to better therapeutic outcomes.
Front pages
xi
Front pages
xii
CHAPTER 1: INTRODUCTION
I began the present study as an interrogation of my own clinical practice. I had been
working for several years as a mental health nurse in the public healthcare system. To
augment my nursing knowledge, I had studied a counselling approach termed
Narrative Therapy, and completed two workshops with Michael White, one of the
originators of the approach. I integrated the approach into my clinical work with people
who were experiencing a range of mental health problems, and I had the sense it was
working. I had also formerly studied literature, and maintained an interest in linguistics.
I was struck by common features between the concept of metonymy advanced within
cognitive linguistics, and a particular strategy within Narrative Therapy, called
externalising the problem, and I began to bring this analytic framework into my clinical
work.
My work role changed, and so did the study. I decided that before I could suggest
changes to clinical practice, it was important to clearly demonstrate that the ideas I was
pursuing had merit outside my head. So I set about investigating what people with lived
experience of mental illness thought. At the same time as I was interested in their
stated ideas, I also wanted to analyse their language using tools from a cognitive
linguistic framework, in order to see if I could identify patterns that may contribute to
deeper understanding of the influence of language patterns.
Language is central to mental health practice and psychiatry. It underpins processes of
assessment, diagnosis, treatment and evaluation. It can influence therapeutic
engagement. At a minimum, it affects the way that people who experience mental
health problems can communicate these to others, and in many cases it is integrally
linked to the experiences themselves. Both ‘everyday’ language, and the particular
language that occurs in mental health practice, depend heavily on people making
inferences, that is, choosing and interpreting words to create shared meaning.
This study is an exploration of the influence of metonymic pragmatic inferencing on
language used by and about people who live with mental illness. It uses knowledge
developed in cognitive linguistics to do this.
Chapter 1 Introduction
1
Key terms
At the outset, I will introduce a few key terms, to clarify how they will be used in the
study, and give the reader a foothold:
Metonymy is a language practice where one term is used to stand for another that is
closely related to it conceptually.
Cognitive linguistics is an approach to the study of language as ‘an instrument for
organising, processing and conveying information’ (Geeraerts 2008, p. 3).
Pragmatics is the study of language in use.
Inferences are conclusions drawn from cues. They ‘are based on knowledge, previous
experience, expectations and needs’ (Stein-Parbury 2014, p. 196).
Aim and Objectives of the study
The aim of this study was to explore the influence of metonymic language on the
experience of people who live with mental illness, including their experience of
treatment.
This was achieved by the following objectives:
1. identifying metonymic practices in the clinical language that people who live
with mental illness encounter and use themselves
2. exploring the significance of identity for people who live with mental illness,
including the influence of diagnosis
3. identifying metonymic language in the speech production and comprehension of
people who live with mental illness
4. assessing if these elements influence their experience of communication with
others, including clinicians.
Research questions
The study had two research questions:
How does the use of metonymic language influence the experience of people who
live with mental illness?
How does metonymy influence the speech production and speech comprehension
of people who live with mental illness?
Chapter 1 Introduction
2
Background
LANGUAGE AND MENTAL ILLNESS
Language plays a central role in mental health practice. Clinical observation of the
language of people who consult mental health services is key to assessing the
problems they are experiencing (Andreasen & Grove 1986). This observation includes
both language production and language comprehension. Within language production,
both the content of what people say, and the form in which they say it are considered,
as difficulties with each aspect contributing to understanding the problem and
formulating plans for response.
The bulk of research on language and mental illness has concentrated on
schizophrenia. The current textbook description of schizophrenia characterises the
illness as comprising three dimensions: positive symptoms such as hallucinations and
delusions, negative symptoms, such as avolition and anhedonia, and disorganisation
(Lewis, Escalona & Keith 2009). The Diagnostic and Statistical Manual of Mental
Disorders DSM5 lists ‘disorganized speech’ as one of the three core diagnostic criteria
for schizophrenia (along with delusions and hallucinations) (American Psychiatric
Association 2013, p. 99). Observations of language use contribute to the diagnosis of
other mental illnesses as well, most notably bipolar disorder (Akiskal 2009).
It is important to acknowledge at the outset that language does not comprise the whole
experience of mental illness. People experience both ordinary and unusual things that
are not put into words, to others or even to themselves. Similarly, clinical observation
does not directly take into account inner speech, or the words that people ‘hear’ in their
heads, or ‘say’ to themselves. This is not accessible to others, except through the
versions of inner speech that people attempt to share. Notwithstanding these
limitations, clinical practice relies heavily on evidence gathered through spoken
language.
Language is also critical to the delivery of treatment for mental illness. The first
interventions when a person presents with mental health problems typically include
psycho-education, which comprises an explanation of the problem in clinical terms, the
evidence for the clinician’s assessment of the issue, and the proposed treatment
approaches. This initial therapeutic communication may be augmented by
pharmacological, technological and behavioural interventions. These too are mediated
through language, with explanations of the desired effects and potential side-effects of
each intervention delivered prior to and throughout these therapeutic strategies.
Chapter 1 Introduction
3
In addition, language is also used to evaluate the effectiveness of treatment. In the first
instance, this is assessed by the person’s own statements about their response,
literally, do they feel better or worse? This information can be corroborated by reports
from other people in the person’s life, with their consent. Clinical observations of
language use conducted at initial assessment are also repeated, formally or informally,
to note if there has been any change. Given the critical role played by language across
all aspects of mental health practice, examining how language itself is currently
understood is warranted.
CONTEMPORARY APPROACHES TO LANGUAGE
This study considered the influence of language on the experience of people with
mental illness. The study used ideas from two contemporary approaches within the
broader discipline of linguistics: cognitive linguistics and pragmatics.
COGNITIVE LINGUISTICS
‘Cognitive linguistics is an approach to the analysis of language that focuses on
language as an instrument for organizing, processing, and conveying information’
(Geeraerts & Cuyckens 2007, p. 3). Commentators and critics alike agree that there is
no standard unified approach that can be defined as cognitive linguistics, but there are
a number of guiding principles. These have been summarised in a ‘rough guide’ to
cognitive linguistics:
x
Language is about meaning.
x
Linguistic meaning is perspectival – Meaning is not just an objective reflection
of the outside world, it is a way of shaping that world.
x
Linguistic meaning is dynamic and flexible – meanings change.
x
Linguistic meaning is encyclopaedic and non-autonomous – it is not separate
from other forms of knowledge of the world that a person has.
x
Linguistic meaning is based on usage and experience (Geeraerts 2008).
The first point is also the major challenge that cognitive linguistics addressed to the
dominant approaches to the study of language throughout the twentieth century, the
structuralist approach introduced by de Saussure (1916/1974), and Generative
Grammar (Chomsky 1988), both of which give primacy to the rules that govern
language use, rather than its role in interpersonal communication. Cognitive
Linguistics asserts that meaning is created both in the traditional domain of word
choice, covered in semantics, but also in the grammar of languages.
Chapter 1 Introduction
4
Geeraerts’ (2008) second point can be seen as a weak version of a concept termed the
Sapir-Whorf hypothesis, that, ‘The language we speak influences the way we think.’
(Kovecses 2006, p. 34). While the work on which this position was based has been
criticised (Alford 1978), the deliberate manipulation of language to achieve desired
ends persists in media, advertising and political discourse. Geeraerts’ (2008) third
point, that meanings change, can be demonstrated both historically, as the meanings
of words change over time, and ‘online’ as inferences are created through the use of
words in specific contexts.
The fourth point, that meaning is encyclopaedic, is also listed as a key principle by
Croft (2009, p. 396):
This principle implies that one cannot separate a subset of semantic features or
predicates as constituting "the meaning" of a word or construction. Instead, all
that the speaker knows about the real world experience denoted by the word or
construction plays a role (however small) in its meaning.
This view can be contrasted with the idea that meaning can be contained within a
dictionary, that is, fixed to a rigid, definable, unchanging set of criteria. This difference
becomes important within the present study, in relation to how inferential meanings are
constructed (Stein-Parbury 2014).
Geeraerts and Cuyckens (2007, pp. 3-4) summarise elements of the cognitive linguistic
approach:
Language, then, is seen as a repository of world knowledge, a structured
collection of meaningful categories that help us deal with new experiences and
store information about old ones. Specifically, language is a way of organizing
knowledge that reflects the needs, interests, and experiences of individuals and
cultures.
The implications of these ideas for a study of language use in mental health practice
are clear. Language allows people to store knowledge about the world, but also
enables them to deal with new experiences. Moreover, language reflects the needs
and experiences of individuals and cultures. Mental health care frequently involves
conflict; particularly when people experiencing psychosis or disorganisation encounter
a mental health system that is designed to assist them, but also incorporates practices
that seek to contain them and control their behaviour. This conflict is enacted through
Chapter 1 Introduction
5
language, and the system’s power stems in part from medical and legal discourses that
may be completely new to the individual in need of help.
PRAGMATICS
Another approach to the study of language has run concurrent to the development of
cognitive linguistics over the last few decades. Pragmatics is the study of language in
use. Proponents have also distanced the approach from the formal approaches of de
Saussure (1916/1974) and Chomsky (1988), but for different reasons than cognitive
linguists. Where cognitive linguistics adopts insights from cognitive science, pragmatics
is an extension of studies in the philosophy of language, specifically speech-act theory.
The reasons for studying language as it is actually used by people are set out by
Verschueren (2009, p. 2):
Talking, or using language expressively and/or communicatively in general,
consists in the continuous making of linguistic choices, consciously or
unconsciously, for linguistic or extra-linguistic reasons… A theory of language
use could and should therefore be conceived of as the study of the mechanisms
and motivations behind any such choices and of the effects they have and/or
are intended to have.
Again, the reasons why these ideas are important in a study of language use in mental
health practice are clear, most obviously in the two binaries Verschueren posits as
governing language choices. In the first instance, language choices may be conscious
or unconscious. In the second, the reasons that motivate these choices may be internal
to language, or related to dynamics external to language. Each of these aspects can
potentially influence the inferences about meaning that are made in communication. A
speaker may or may not be conscious of the impact their choice of a particular word
will have on a listener. The inference a listener makes on hearing a term may relate to
knowledge they have about the world of which the speaker is unaware. The
applicability of these ideas in mental health practice relate to the range of influences on
the inferences participants make in clinical communication.
In recent years, practitioners in both cognitive linguistics and pragmatics have noted
that there were several areas of mutual interest in their approaches. Cognitive
linguistics shared an interest in how linguistic meaning ‘is based on usage and
experience’ (Geeraerts 2008, p. 5). Equally, pragmatics was alert to motivations and
constraints on language use, which are cognitive processes. Cognitive pragmatics is a
recent theoretical enterprise that reflects the similar objects of attention in language
Chapter 1 Introduction
6
that characterise cognitive linguistics and pragmatics, notwithstanding the different
theoretical provenance of the two disciplines. Schmid (2012, p. 3) offers a definition of
the emerging field:
Cognitive pragmatics focuses on the cognitive aspects of the construal of
meaning-in-context. This pertains to both language production and
comprehension, and it specifically concerns one of the key questions that
pragmatics has set out to answer: What are the cognitive abilities and
processes required to be able to arrive at “what can or must be said” in order to
get across “what is meant” and to arrive at “what is meant” on the basis of “what
is said”.
Cognitive pragmatics sets out a deliberate approach to answer pragmatic questions
using cognitive means. Moreover it demonstrates significant parallels with mental
health practice: mental health workers make decisions about what people mean, ‘on
the basis of “what is said”’, and must use their ‘cognitive abilities and processes’ in
order to effectively convey therapeutic messages.
METONYMY
Etymologically metonymy means ‘change of name’. Initially defined two thousand years
ago as a figure of speech used in rhetoric (Al-Sharafi 2004), metonymy has come
under renewed scrutiny in the last three decades within the discipline of cognitive
linguistics, in which metonymy is viewed as strongly contributing to everyday language
(Lakoff & Johnson 1980), and having powerful influence on inferences that are drawn
in communication.
The title for my study is taken from the title of a book of essays, Metonymy and
Pragmatic Inferencing edited by Panther and Thornburg (2003b). This volume was an
early entry in the Cognitive pragmatic literature. In their introduction, the authors note
that:
Metonymies may be called natural inference schemas, i.e. easily activatable
associations among concepts that can be used for inferential purposes (Panther
& Thornburg 2003a, p. 8).
There tends to be more agreement about how metonymy functions pragmatically, than
on how it is defined. Two problems confront a researcher attempting to provide a
straightforward cognitive linguistic definition of metonymy in a study conducted outside
the field of cognitive linguistics itself. The first is that, although cognitive linguists have
Chapter 1 Introduction
7
many interesting things to say about metonymy, they don’t necessarily agree with each
other, particularly when it comes to defining the concept. The second is that definitions
frequently utilise terms that have specific meanings within cognitive linguistics that
aren’t shared in the broader community, for example, ‘source’, ‘target’ and ‘domain’.
These issues will be discussed in greater depth in Chapter 3: Metonymy. For now, I will
adopt a suggestive attempt at definition of the term offered by Croft and Cruse (2004,
p. 48).
Metonymy is, loosely, the use of a word to denote a concept other than its
'literal' denotation. A cognitive linguistic analysis of metonymy is the ability of a
speaker to select a different contextually salient concept profile in a domain or
domain matrix than the one usually symbolised by the word.
In selecting a particular source (‘concept profile’ above) as salient in context, speakers
draw the listener’s attention to the targeted meaning, and this practice is particularly
effective in making communication economic. For example, in the sentence ‘Canberra
announces policy changes’, the targeted meaning is ‘the Australian Federal
Government based in Canberra’, but the source is simply the name of the city. That the
speaker intends some form of governing body is inferred from the information that
follows in the sentence itself, that is, the language-internal context. The particular
source selected also effectively distinguishes which government is intended as the
reference, through knowledge external to language, that is, which arm of government
operates from this location. For most listeners, the correct pragmatic inference will be
drawn from the use of the term ‘Canberra’ in this context, with less cognitive effort than
processing ‘the Australian Federal Government based in Canberra’ would demand. Of
necessity, the selection of one attribute of an entity or concept for attention relegates
other aspects to the background (Langacker 1993; Talmy 2007). Conversely, attributes
of the selected source can carry over into the target meaning, for instance,
longstanding ideas about the city of Canberra that a listener holds. Both of these
processes can lead to inferential effects that are not immediately obvious or
predictable.
METONYMY AND MENTAL ILLNESS
Referring to a person by their diagnosis (e.g., ‘a schizophrenic’) is an example of
metonymic linguistic practice; the diagnosis is selected as the salient attribute, and is
used to stand for the whole person. It is a commonplace practice, and does not occur
exclusively in the clinical settings where such salience is apposite, but can also operate
in the home, the workplace and in social and other media. The metonymic use of
Chapter 1 Introduction
8
mental health diagnoses as labels to stand for the people who experience mental
illness contributes to stigma and depersonalisation. A recent survey conducted in
Australia revealed the enduring prevalence of stigmatising attitudes toward people with
mental illness, particularly those diagnosed with schizophrenia (Reavley & Jorm 2011).
People who live with mental illness are not exempt from these stigmatising attitudes,
and the problem of ‘self-stigma’ is associated with low self-esteem and poor outcomes
(Corrigan, Kosyluk & Rusch 2013; Lysaker, Roe & Yanos 2006).
Beyond its role in underpinning the practice of stigmatising labelling, metonymy has a
neglected history within psychiatry. Researchers in the 1930s and 1940s noted unusual
speech patterns used by some people diagnosed with schizophrenia, which they linked
to metonymy (Cameron 1944; Goldstein 1944). This was characterised as the use of
idiosyncratically selected attributes to refer to things. It was regarded as inhibiting
effective communication, and indicative of underlying thought disorder. The concept of
‘metonymic speech’ has received only intermittent attention in the following decades
(Andreasen 1979a; Sadock 2009), and has not informed therapeutic approaches.
METONYMY AND THERAPEUTIC APPROACHES
It is in therapeutic approaches to managing stigma that the unnamed but implicit role of
metonymy has been taken up. Stigma about mental illness remains prevalent,
notwithstanding decades of work to reduce this. One way in which effective
approaches work is by recreating a conceptual distance between the diagnosis or
label, and the person to whom it is applied. This enacts a form of reverse-metonymy,
countering the process by which a diagnosis is seen to ‘stand for’ the person. Given the
ubiquity of metonymic referencing in everyday speech, addressing the power of
stigmatising language is a complex undertaking, but it is a key aim for many people
who live with mental illness and their advocates.
One approach that directly addresses stigma as part of therapeutic intervention is
Narrative Therapy (White & Epston 1990). Drawing on sociological writings about
stigma and marked identities (Goffman 1963), and philosophical work on the
construction of modern identity (Foucault 1980), therapists working with this approach
developed practices that extended beyond the focus on the individual in traditional
psychotherapies. A key observation was that many people struggling with a range of
mental health problems were hampered by modern notions of identity that link a
person’s whole identity to a specific attribute, often couched in vague terms such as
‘burden’ or ‘worry’. When the selected attribute is somehow marked as negative, there
is frequently a carry-over effect that the person has no capacity independent of the
Chapter 1 Introduction
9
problem that they could potentially summon in dealing with the problem. Narrative
Therapy maps sophisticated strategies for systematically addressing stigmatising
language; the process whereby an attribute stands for a person’s identity matches the
cognitive linguistic definition of metonymy, and thus the strategies to remedy the
process necessarily draw on recognition of metonymy’s influence, though this has
never been explicitly named in the key texts (Bird 2000; White 2007; White & Epston
1990).
THE EXPERIENCE OF METONYMIC LANGUAGE FOR PEOPLE WITH
MENTAL ILLNESS
The metonymic transfer of meaning whereby a diagnosis of mental illness is selected
as the salient attribute in their identity is an example of the power of inferencing. This
holds whether the person adopts such an identity themselves, or if they find it imposed
on them through stigma.
Less is known about the experience of language disorder in mental illness.
Psychiatrists and psycholinguists have demonstrated the occurrence of language
disorder, and conducted experimental studies of various patterns of idiosyncratic or
disordered language use. However, there is little available evidence of what it is like to
experience language disorder, as the majority of attention has been on the lived
experience of positive psychotic symptoms, such as hallucinations and delusions.
Consequently, it is not known whether therapeutic approaches that incorporate
recognition of metonymic language used by individuals with mental illness, the
clinicians who treat them and the broader community that interacts with them can
influence these experiences.
This thesis is an examination of a number of these theories about language and mental
illness, with one critical difference. Alongside an examination of the presence of many
of the linguistic markers identified by previous researchers, the study also elicits the
experience of using, and being subject to language, through the actual words of people
who live with mental illness.
Chapter 1 Introduction
10
Study overview
This study consists of an exploration of the research questions, through nineteen
interviews with ten participants who have a lived experience of mental illness that
includes psychosis. Interviews were analysed thematically, to elicit participant’s views
on their experience. They were also analysed linguistically, to trace the presence of
metonymy and other distinctive patterns in the language used by participants.
Chapter 1: the Introduction, sets out the aims and objectives of the research, and gives
a brief overview and background to the key concepts in mental health practice and
contemporary linguistics that underpin the study.
Chapter 2: Language and mental illness, examines in depth the relationship between
the ideas being explored, and the various literatures these ideas arise from. This
entails consideration of nursing, psychiatric, and counselling literature.
Chapter 3: Metonymy, addresses metonymy primarily from the perspective of cognitive
linguistics.
Chapter 4: Method, details the specific methods used to collect and analyse data. The
method combines elements from more than one research paradigm, and the reasons
for methodological choices are described in detail.
Chapter 5: Results, reports on the analysis of the date provided by participants in the
interviews.
Chapter 6: Discussion, examines how the results of the analysis of the data collected
fits in with the questions asked in the Introduction and the previous literature discussed
in the background chapters. This chapter also addresses the limitations of the present
study and concludes with a consideration of the implications of the study for research
and practice.
Chapter 1 Introduction
11
Chapter 1 Introduction
12
CHAPTER 2: LANGUAGE AND MENTAL ILLNESS
Introduction
This chapter considers the relationship of language and mental illness. In the first
section, the role of language in contemporary psychiatric practice is summarised. This
is followed by a review of historical approaches to the role of language in
conceptualising mental illness. Other approaches to researching and treating mental
illness, including mental health nursing and psycholinguistics, are then addressed.
Language is central to conceptualising and treating mental illness in contemporary
mental health practice. This includes the processes of assessment and diagnosis, the
delivery of treatment, and evaluation of the outcomes of treatment, from the point of the
view of the person with the problem, and the person treating them. The importance of
language is present across various disciplinary practices in mental health, including
psychiatry, mental health nursing and counselling. The chapter critically engages with
ideas from the different disciplines, in order to establish the rationale for the current
study.
Diagnosis
Mental health diagnoses are made through language, with clinicians assessing the
experiences that people describe, and the way that they communicate these
experiences. These clinical observations are formulated into a diagnosis through
matching them to categories defined in language in diagnostic manuals (American
Psychiatric Association 2013). It remains the fact that physiological measures, typically
used in medical diagnosis, are of minimal use in diagnosing mental illness as,
‘currently, there are no radiological, laboratory, or psychometric tests for the
disorder(s)’ (American Psychiatric Association 2013, p. 101). Techniques provided
through brain imaging technology can contribute to understanding of schizophrenia and
other conditions, but they are only used negatively in clinical practice, that is to rule out
the possibility that the presenting symptoms arise from an organic process.
SPEECH PRODUCTION: OBSERVATION AND INFERENCE
Serious mental illness is thus diagnosed principally through language, in particular
through observations of the presenting person’s speech production. The predominant
research on language and mental illness has been conducted on schizophrenia. This is
reflected in the prominence of speech disorder in the diagnostic criteria for the
Chapter 2 Language and mental illness
13
condition. The latest version of the Diagnostic and Statistical Manual of Mental
Disorders DSM 5 includes ‘disorganized speech’ as one of three characteristic
symptoms, along with delusions and hallucinations, that must be present for a
diagnosis of schizophrenia to be made (American Psychiatric Association 2013). The
DSM 5 specifies that:
Disorganized thinking (formal thought disorder) is typically inferred from the
individual’s speech… Because mildly disorganized speech is common and
nonspecific, the symptom must be severe enough to substantially impair
effective communication (American Psychiatric Association 2013, p. 88, italics
in original).
There is considerable room for variation in the application of this diagnostic criterion.
The measure used to determine that observed disorganised speech constitutes a
symptom is the psychiatrist’s assessment that it has led to substantially impaired
communication. This carries the presumption that any communication failure rests with
the individual under assessment, and not with the practitioner. This presumption is
inconsistent with how communication operates dynamically, by definition, between two
or more people (Jakobson, Pomorska & Rudy 1987). The pragmatic understanding that
inferences may be conscious or unconscious, and related to external influences
(Verschueren 2009) is also absent, notwithstanding the reliance on inference to make
the diagnosis.
Disordered speech production has been further differentiated into aspects
demonstrating positive and negative thought disorder:
Positive thought disorder includes tangentiality, derailment, neologisms and
several other phenomena that appeared to be highly correlated in
patients...Positive thought disorder is now generally conceptualized as part of
the disorganization sub syndrome of schizophrenia. It is also termed
disorganized speech (Kuperberg & Caplan 2003, p. 447).
The delineation of the disorganisation sub syndrome is in line with the dimensional
approach to assessing mental illness, which focuses on the particular symptoms
experienced by people, rather than which diagnostic category these symptoms place
them in. This approach is intended to address some of the limits of the categorical
approach, such as difficulties drawing the boundaries between certain diagnoses
(American Psychiatric Association 2013). The dimensional approach extends the
earlier division between positive and negative symptoms by adding the disorganisation,
Chapter 2 Language and mental illness
14
or cognitive impairment sub syndrome, and also recognising the presence of mood
symptoms in formulating treatment plans. In this model, negative thought disorder
which includes poverty of speech, and poverty of content of speech, is considered as
part of the negative sub syndrome of the illness (Kuperberg & Caplan 2003).
Formal aspects of produced speech as markers of thought processes are also used in
the diagnosis of bipolar disorder:
Thinking processes are accelerated, subjectively experienced as flight of ideas,
and thinking and perception are unusually sharp. The patient may speak with
such pressure that associations are difficult to follow; such "clang" associations
are often based on rhyming or chance perceptions and can be lightning fast.
The pressure to speak may continue despite the development of hoarseness
(Akiskal 2009, p. 1704).
In order to describe the presentation of thought in bipolar disorder, the author uses
figurative language, ‘flight of ideas’ and ‘lightning fast’, and inexact measures to
determine pathology, ‘unusually sharp’. The clinician can determine that a perception is
‘chance’ without knowledge of the other person’s inferential schema.
It is clear from these examples that a critical element of the diagnostic process is
reliance on the inferencing practised by clinicians. Moreover, these clinical approaches
typically sidestep the longstanding controversy around the relation of speech to
thought, opting for a general acceptance that thought is relatively transparently
conveyed through speech. Andreasen and Grove (1986, p. 348) make explicit the
short-cut taken in clinical practice:
While it is also possible to obtain indices of cognitive performance or 'thought'
using formal tests to elicit disordered thinking, such as proverb interpretation or
projective tests, in a clinical setting we usually infer a person's thoughts directly
from his speech.
A contemporary description of this inferential process introduces an analogy equating
psychiatric observation of language with technological approaches used elsewhere in
health care:
As a primary care physician uses the fundoscopic examination as a way to
visualize the vasculature of the central nervous system (CNS), in psychiatry
speech is examined as a means to determine a patient's verbal cognition
(Lewis, Escalona & Keith 2009, p. 1448).
Chapter 2 Language and mental illness
15
The observation of language to infer the presence of thought disorder and diagnose a
person with mental illness is a significant clinical responsibility. The diagnostic criteria
outlined above are drawn from a diagnostic manual and a key textbook, both used as
references by practitioners, and as tools to prepare future mental health workers. The
language they use is figurative, inexact and inconsistent with current theories of how
language operates.
Compare Hobbes (1650/2011, p. 49), writing over 300 years ago, describing his own
thought patterns, which include a seemingly chance perception arrived at through quick
but coherent thought:
In a discourse of our present civil war, what could seem more impertinent than to
ask (as one did) what was the value of a Roman penny? Yet the coherence to
me was manifest enough. For the thought of the war introduced the thought of
the delivering up the King to his enemies; the thought of that brought in the
thought of the delivering up of Christ; and that again the thought of the thirty
pence, which was the price of that treason: and thence easily followed that
malicious question; and all this in a moment of time; for thought is quick.
There is, if anything, less figurative language in Hobbes’ (1650/2011) account than in
the diagnostic criteria previously described. He also highlights how negative inferences
made about instances of speech (‘impertinent’, ‘malicious’) can miss their underlying
coherence.
Variation in application of diagnostic criteria has been observed occurring:
Certainly, different societies, different individuals and even different clinicians
have a variety of thresholds for both noticing atypicality and for labelling it
pathological (Fine 2006, p. 300).
The inexact and under specific language in the current written tools used to support
and develop clinicians to effectively diagnose mental illness does not adequately
protect against such variation. This has implications for the people who are diagnosed
with mental illnesses.
SPEECH PRODUCTION: CONTENT AND FORM
In the clinical observation of speech production, a division is made between the content
of what people say, and the form in which they say it. The content of speech is a key
element of the clinical assessment process. The person may describe thoughts or
emotions that are different to their previous thoughts and emotions, and are
Chapter 2 Language and mental illness
16
unwelcome. They may describe beliefs about things that are not shared with anyone
else, are not consistent with external evidence, and evince dramatic shifts from their
previously held beliefs (delusions). They may describe the experience of hearing or
otherwise sensing things that no one else can hear (hallucinations). They may explicitly
describe what they hear. They may describe their mood, and their experience of
alterations in mood. They may describe changes in thinking that are congruent with
these moods or not. The terms they use for these experiences may be drawn from a
range of available discourses, including but not limited to psychiatry. Thus a person
may report that they experience auditory hallucination, or they hear voices, or they hear
someone talking to them. Each of these terms, while possibly describing the same
phenomenon, carries different inferences.
Behavioural manifestations of altered reality must also be described in language to
form part of the clinician’s formulation of the problem. In the situations described
above, it is the content of the person’s speech that contributes to the formulation of the
problem, and to this extent, the pattern of communication between person and
therapist is reasonably straightforward. In assessments, clinicians need to be sensitive
to cultural differences in beliefs, for example, in numerous cultures it is not considered
abnormal to believe that the voices of one’s ancestors speak directly to people. In other
cultures it is typical for mood to be communicated in somatic terms. Content of speech
can often be corroborated by information from others, though this is not always
undertaken.1
The form of speech refers to the manner in which the person speaks, distinct from the
content of what they say. Observations of speech form can include the rate of speech,
the typicality of constructions, the recurrence of terms, and the congruence to the
current topic. These alterations in the formal aspects of speech range from mild to
severe. Mild alterations can seem little different to the idiosyncratic speech patterns of
all speakers, for instance, the habit of circling around a topic (circumstantiality), or
answering questions in tangential or elliptic ways.
In order to more clearly define the formal deviations in speech that could be used in
formulating differential diagnoses of mental illness, Andreasen (1979a, 1979b)
developed the Thought, Language and Communication Scale. This work built on earlier
A colleague with a name that is unusual in the Australian context once discovered a person
she cared for had been hospitalised in a different area. She phoned to enquire about the
person’s welfare, and when she identified herself, the clinician paused then stated, ‘Oh, you’re
real’. From lack of cultural familiarity with the worker’s name, the treating team had inferred she
was a delusional construct of the person they had detained.
1
Chapter 2 Language and mental illness
17
descriptions of commonly observed patterns of language, including terms used by
Kraepelin (1919/1999) and Bleuler (1911/1950). Eighteen different patterns of formal
speech disorder are described, with each given specific criteria, examples, and
important exclusions. Metonymy is included in the category of ‘word approximations’
(Andreasen 1979a). Further work used sub-sets of observed patterns of speech to
differentiate between sub-syndromes in schizophrenia, and determine prognostic
features (Andreasen & Grove 1986).
While Andreasen’s (1979a, 1979b) work has continued to be used in research contexts
(see e.g., Docherty 2012), it is not included in teaching and reference materials, except
in diluted forms. Current textbooks typically include long lists of signs of formal speech
disorder, but these are defined very briefly, with minimal comment about the relative
frequency of these patterns, or indications about what they may suggest in terms of
differential diagnosis. For instance tangentiality and derailment are both defined as
moving away from the topic, but the key differential element, that tangentiality is in
response to a question, while derailment occurs within a person’s own speech, is not
included (Sadock 2009).
SPEECH COMPREHENSION
Speech comprehension has traditionally been given less attention than speech
production in the diagnostic process (Kuperberg 2010a). Some people with
schizophrenia experience particular difficulty comprehending the speech of others. This
can relate to specific aspects of speech, like interpretation of figurative language, or
idioms. Or it can relate to following the course of a conversation when the meaning
they understand a word to have does not fit with the context, and they cannot correctly
interpret the intended inference.
Asking people to explain common proverbs, for example, ‘A stitch in time saves nine’,
previously formed a standard part of diagnostic procedures, used to determine if
people could think in abstract ways, or were constrained to the concrete meanings of
terms. This practice has reduced in recent times, as clinicians became more aware of
the cultural biases that inform many proverbs, leading to incorrect conclusions being
drawn based on proverb tests. Nonetheless, for certain subgroups of people with
schizophrenia with shared cultural backgrounds, researchers have observed differential
patterns in interpretation of speech, particularly in interpretations of figurative language
and indirect requests (Champagne-Lavau & Stip 2010; Chapman 1960).
Chapter 2 Language and mental illness
18
For people experiencing paranoia, speech comprehension can be characterised by
heightened vigilance, with seemingly unloaded terms being recognised by the
individual as specifically referential within their schema. This is not restricted to
linguistic messages. People have reported ‘reading’ specific messages intended for
them in the paralinguistic form of vehicle number plates and even in the colours of
passing cars.
DIAGNOSTIC HETEROGENEITY
A further complication in formulating mental health diagnoses is that each diagnosis
can be applied to a heterogeneous group of symptoms. Using the DSM 5 criteria
(American Psychiatric Association 2013), one person may be diagnosed with
schizophrenia who has experienced delusions, but never experienced a hallucination,
while another may have had mild voices and a history of disorganised behaviour, but
never had a delusion. This disparity is not clear to the people who receive the
diagnosis, and may lead to confusion, particularly if they don’t have symptoms that
form part of the cultural image of the illness to which they had been previously
exposed. For instance, people who have never experienced hallucinations may query
why they have been diagnosed with schizophrenia, and others who have never
experienced depression question how their mood can be called ‘bipolar’.
Equally, symptoms can occur across a number of diagnoses. A person who has
experienced hearing voices may find themselves diagnosed with schizophrenia,
schizoaffective disorder or bipolar disorder, depending on the co-presence and timing
of other symptoms related to changes in their mood. This variation can easily lead to
confusion in people and their family members grappling with their entry into the mental
health care system.
FIGURATIVE LANGUAGE
Figurative language is prevalent in the clinical descriptions of language disorder in the
diagnosis of mental health. Examples from the diagnostic literature include, ‘flight of
ideas’ and ‘lightning fast’ (Akiskal 2009, p. 1704), ‘derailment’ and ‘tangentiality’
(Sadock 2009). The use of figurative terms to describe phenomena observed in mental
illness dates back to the early days of psychiatry. Eugen Bleuler coined the term
‘schizophrenia’ in 1908. Fusar-Poli and Politi (2008, p. 1407) describe the process:
Bleuler collected material directly from his passionate clinical work. By
accommodating himself to the spatial and temporal environment of his patients,
he realized that the condition was not a single disease…was not invariably
Chapter 2 Language and mental illness
19
incurable, and did not always progress to full dementia… the splitting of
different psychological functions, resulting in a loss of unity of personality, was
the most important sign of disease in Bleuler’s conception. Thus, he challenged
the accepted wisdom of the time and advanced his purportedly less static and
stigmatizing concept by juxtaposing the Greek roots schizen (to split) and phren
(originally denoting “diaphragm” but later changing, by metonymy, to “soul,
spirit, mind”) (italics and parentheses in original).
It is ironic that the term schizophrenia was deliberately constructed to counter a
stigmatising conception inherent in the previous term dementia praecox, but that name
has subsequently accrued a heavy burden of stigma in popular consciousness.
Bleuler’s (1911/1950) deliberate use of figurative terms in developing a new conception
of the illness was linguistically transparent.
The use of figurative language in psychiatry is still prevalent. As discussed above,
published definitions tend to be underspecific. From a cognitive linguistic perspective,
the source of each term would not be sufficient to allow clear access to the target
meaning, and could be intrusive to the point of misleading. For instance, ‘derailment’
brings to mind a train coming off the rails and stopping, catastrophically or otherwise,
but the language pattern the term describes in psychiatry in fact does not necessarily
entail that speech comes to a halt. This can affect the communicability of clinically
observed signs, particularly between mental health workers and the people actually
experiencing these symptoms of mental illness.
The use of figurative language in clinical communication has been specifically
addressed by a few authors (Crawford et al. 1999; Hamilton & Manias 2006). Though
these articles focus more on how mental health workers communicate to each other
about people with mental illness, they do trace the potential influence of this on
clinician’s direct communications with the people they treat. One recent Australian
textbook advises against any use of figurative language altogether, due to the potential
for inferences to be confusing across cultural lines (Candlin 2008), however this
recommendation is not accompanied by pragmatic suggestions for communicating
without figurative language.
Delivery of treatment
Therapy for mental illness is multidimensional: treatment comprises psychological,
pharmacological, technological and behavioural interventions. Many mental health
professionals use more than one treatment approach in their work, either as sole
Chapter 2 Language and mental illness
20
practitioners, or collaboratively with members of other health disciplines. Thus a person
may see a medical professional who prescribes medication and monitors its effects,
and a nurse or other qualified mental health professional, who engages in counselling.
Alternately, a psychiatrist could function as both prescriber and counsellor. Even in the
rare cases where treatment is exclusively pharmacological, formulation of the
diagnosis, explanation of the proposed treatment, and monitoring of its effectiveness all
proceed via language, and are therefore mediated by language.
An important element in the delivery of treatment is interpersonal communication.
Mental health workers communicate their ideas to the people who consult them about
mental health problems. Moreover, workers communicate to each other. In doing so,
the choice of language used can set up inferences that extend beyond the explicit.
There are gaps between the types of language used by different disciplines.
There is extensive literature on effective therapeutic communication (see, e.g. Peplau
1952/1988; Stein-Parbury 2014) however, notwithstanding the comprehensive attention
thought disorder and communication difficulties have received, this has not translated
into interventions directly targeting problems with speech comprehension or production.
Bowie and Harvey demonstrated a positive correlation between communication
difficulties and poor social outcomes for a population with schizophrenia (Bowie &
Harvey 2008). This finding was confirmed by Tan and colleagues, in a younger
population living in the community (Tan, Thomas & Rossell 2014). These authors
recommend that rehabilitation include focus on improving communication, but there is
limited literature on implementation of this approach (Holshausen 2012). Studies have
reported the effects of social skills training (Mueser et al. 2010) and cognitive
remediation (McGurk et al. 2007), but to date neither of these approaches have
explicitly targeted language skills development.
Only one study was located that specifically addressed language difficulties in
schizophrenia from a speech therapy approach (Clegg et al. 2007). Unfortunately, the
intervention, though deemed to have some benefit, involved a team engaging in weekly
speech therapy sessions with one person throughout the course of a nine-month
inpatient admission, and is thus not practicable in clinical services. Interventions
targeting the processing of figurative language have been reported in a population of
people with autism who experience related communication difficulties (see, e.g.
Melogno & Pinto 2014), however the literature does not yet include evaluation of the
effects.
Chapter 2 Language and mental illness
21
Evaluation of treatment
Evaluation of the outcomes of mental health practice is also conducted through
language. In the first instance, the person may report feeling better. This can be a
report of the absence or quieting of hallucinations, or shifts in disturbing beliefs, or a
sense of stability in mood. These changes can be accompanied by evidence from other
aspects of the person’s life, for example, better relationships with other people in their
life, and/or return to work or study. The person’s report can equally be corroborated by
reports from others in a position to provide useful information, such as family members.
In cases where the problem has manifested in disorder of speech, observed changes
in speech production can be used to evaluate treatment. If a person whose speech
production was generally incomprehensible to the intended audience develops the
capacity to communicate more clearly, this is interpreted as a sign of clinical recovery.
In extreme cases this is relatively easy to assess.
Such an assessment of recovery leaves an unresolved question. There are no routine
measures of normal speech production, and idiosyncratic patterns of speech can be
argued to be completely unrelated to mental health problems. Nor are the measures by
which a person’s speech is deemed unintelligible or opaque clearly agreed upon.
Evaluation of clinical recovery therefore includes variation, and subsequently a
significant margin of error.
Observations can be formalised in outcome measures, which are tools used to
evaluate the effectiveness of interventions at a population level as well as for the
individual (see, e.g. Rosen, Hadzi-Pavlovic & Parker 1989; Wing et al. 1998). These
measures have an increasing importance in the kind of services that are made
available to people. Items are rated on scales, and reported as quantitative data, but
the initial items are described, and subsequently rated, in language. Lewis, Escalona
and Keith (2009) have identified the risk attendant on the use of such scales:
But the existence and reliable utility of these scales has nonetheless had the
unintended effect of focusing researchers' and clinicians' attention on the
questions asked by the scales and diverting some attention away from the
stories patients have to tell. The scales, like the brief and reliable contemporary
diagnostic criteria we apply, have become the illness in the mind of many in
mental health, and so the individual experience of patients suffering with the
illness does not always have a place in clinical interviews.
Chapter 2 Language and mental illness
22
This observation suggests a metonymic influence on researchers and clinicians
attention, whereby the measures used to report outcomes come to stand for the
experiences of the people they treat.
Medico-legal implications
If a person is determined by psychiatrists to have a mental illness that requires
treatment in order to prevent harm occurring to the person themselves, or to others,
they can apply for a legal order that directs the person to accept the prescribed
treatment whether they consent or not. These directions can set out the physical
location in which a person is treated, to the extent that a person may find they are
involuntarily confined in a hospital for a set period of time. This is experienced by many
as a deprivation of liberty akin to being imprisoned as punishment for a crime. People
can also be required to accept treatment with medications that change the way they
feel, and result in side-effects, such as sedation and weight gain, that alter their
experience, and potentially their sense of identity.
These processes occur through language, and introduce a further discourse, legal
language, to the already unfamiliar medical discourse framing the materiality of the
person’s experience. In Australia, mental health legislation is established and enacted
at a jurisdictional level, that is, by each of the states and territories. In the absence of
consistent national mental health legislation, people can discover that they have
different rights in different jurisdictions. Jurisdictions periodically review their mental
health legislation, and this typically involves changes in the language used, in particular
to reflect evolving principles about the capacity of people with mental illness to engage
in decision making about their treatment (see, e.g. Mental Health Act (Victoria) 2014;
Mental Health Act (Western Australia) 2014).
The possibility that a person may become subject to a legislative order that allows for
treatment to be provided coercively introduces a further consequence to clinical
assessment of speech:
There can be tendencies, and more or less legitimate reasons, to keep silent,
not to convey what one saw, did or experienced (Fischer-Rosenthal 2000, p.
119).
Fischer-Rosenthal (2000) is an historian whose work included interviewing German
people about their actions during World War II. However, his statement that people
may have what they regard as legitimate reasons not to disclose their experience
Chapter 2 Language and mental illness
23
resonates with the situation of a person undergoing a mental health assessment with
the prospect that they may be detained against their will in a mental health unit. In a
clinical observation, Andreasen (1979a, p. 1316) noted that:
Anyone can exert conscious control over his language behaviour and
manipulate it in various ways to conceal or obscure his thoughts.
Other authors have noted that a person’s use of language may not be deliberately
deceptive, but nonetheless obscure. Where Andreasen (1979a) highlighted conscious
determination to avoid transparent communication, Peplau (1952/1988, p. 197)
identified psychological reasons that can interfere with people expressing themselves
clearly:
The patient cannot be forthright about expressing the meaning of events and so
often uses a highly disguised way of saying what is felt. Nurses can speculate
on the meaning and often come close to what it actually means to the patient.
This is a clear statement that effective listening involves nurses (and, by extension,
other mental health workers) practising skilled inferencing. Menzies Lyth (1989, p. 28)
asserts a more radical psychoanalytic position:
It is obvious that people do not say what they really mean even when they
honestly and sincerely say what they consciously think, let alone when they do
not.
This invokes the famous title of an influential essay by the language philosopher Cavell
(1958), ‘Must we mean what we say?’ which examines the gap between semantic and
pragmatic meaning and the consequent inferences and implications that arise in
language use. Cavell (1958, p. 181) states that:
Learning what these implications are is part of learning the language; no less a
part than learning its syntax, or learning what it is to which terms apply: they are
an essential part of what we communicate when we talk (italics in original).
The complexities of communication in general that have been identified by these
authors do not figure in the way that language is portrayed in psychiatric textbooks
(e.g., Sadock 2009) and diagnostic manuals (American Psychiatric Association 2013).
Similarly, the suite of mandatory assessment documentation to be completed by public
mental health workers in NSW does contain an item for ‘Communication Issues’ but the
suggested prompts are limited to ‘language or cultural barriers, sensory impairment’
Chapter 2 Language and mental illness
24
(Mental Health and Drug and Alcohol Office 2008). Clinicians are therefore not
supported to consider the influence of linguistic features in the inferences they draw
and respond to, either in their training or in their practice. This has implications for their
medico-legal decision making, which in turn has implications for people who live with
mental illness.
Historical approaches to language and mental illness
This section presents a selection of earlier approaches to language and mental illness.
EARLY TWENTIETH CENTURY APPROACHES
Psychiatrists from the beginning of the twentieth century have studied how language
provides information about mental illness. Bleuler (1911/1950, p. 150) gave numerous
examples of the language of his patients, including this metonymic substitution:
Sometimes the figures of speech misuse the principle of pars pro toto, in such a
way that its least essential component is selected to represent the total concept.
For example, a shoe is called "something used to dance in".
Bleuler (1911/1950) uses a Latin phrase, ‘pars pro toto’ which translates as part for
whole, that is, one of the classic forms of metonymy. While noting the phenomenon, he
did not refer directly to ‘metonymic speech’. Blueler (1911/1950, p. 150) further notes
that:
Frequently, the similarity of concepts which leads to the interchange of words is
an extremely tenuous one and involves thought processes which have no
connection with the thought being expressed… Sometimes the similarity is not
in the concepts but rather in the words, this can result in dull plays on words.
Many of Bleuler’s (1911/1950) ideas have been taken up by later researchers,
including the phenomenon he described above, where word use can be prompted not
by context, but by associations with the sound or meaning of the actual words, which
would later be developed into the concept of semantic priming. The language evidence
Bleuler (1911/1950) reported arose from the speech of patients in his care. He focused
on their speech production, rather than their comprehension. The context for their
speech is not explicitly described, but can be understood to occur in clinical
discussions between patient and psychiatrist. Bleuler was alert to the fact that speech
does not exactly mirror thought and that shifts can occur at the level of words only, but
also between concepts and words.
Chapter 2 Language and mental illness
25
Where Bleuler (1911/1950) observed deviations in figurative speech production in his
patients, Vygotsky (1934/1994, p. 321) observed difficulties they had in
comprehension:
Most remarkable was that I found disturbances in the understanding of words
figuratively used, even when there was no apparent disturbance of intellectual
life in general. This difficulty became very obvious when special words or
concepts were used. While the normal mind has no difficulty in using given
words metaphorically or figuratively, the same problem presents
insurmountable difficulty for the patient with schizophrenia in spite of the fact
that he has retained from childhood the habit of using figures of speech,
proverbs, etc.
Much of Vygotsky’s (1934/1994; 1934/1986) work concentrated on the acquisition and
development of language in children, and he spent time exploring the theory that
schizophrenia represented a regression to a less sophisticated way of thinking, made
manifest as an incapacity to process figurative language. Cameron (1938) studied this
theory experimentally, and concluded that regression was not an accurate model for
the characteristic speech observed in people with schizophrenia. The theory of
schizophrenia as regression was abandoned at this time.
1930S AND 1940S: METONYMIC DISTORTION
During the 1930s several psychiatric researchers were pursuing studies on the
relationship between language and mental illness. The pattern of speech production
previously observed by Bleuler (1911/1950) was defined as metonymic by Cameron
(1938, p. 20), drawing on traditional rhetoric:
Metonymic distortion consists of the substitution of an approximated but related
term or phrase for the more precise definitive term that normal adults would
presumably use in the same setting (italics in original).
In a later study, he writes of the effect of this type of speech on the listener, and
expands on how difficult it can be to interpret:
The result is analogous to what one gets in looking at something through
steamed glasses. One patient says that he "has menu three times a day"
instead of food or meals. Taken in the frame of the individual patient's life and
his known fantasies, these sentences can be translated into more precise,
socially current forms. It is this need for continual translation and sharpening of
Chapter 2 Language and mental illness
26
the focus that confuses and fatigues the listener. It gives the same feeling of
strain that comes when one tries to follow a conversation in a foreign tongue
with which one has some acquaintance but not quite enough. One goes along
for a bit all right, but then begins to slip behind and miss the meaning (Cameron
1944, p. 54).
Cameron (1944) identifies that the terms chosen, while seemingly idiosyncratic and
superficially opaque, can frequently be correctly inferred when the broader context of
the speaker’s language practice is understood. This is another example where the
writer resorts to figurative language, here analogy, to convey an aspect of clinically
observed language, in this instance comparing comprehending the speech of a person
with schizophrenia to understanding someone speaking in a foreign language and to
looking through steamed glasses.
Cameron (1944, p. 54) is not immune to a particular irritation that characterises
responses (see, e.g. Bleuler 1911/1950; Chapman 1960) to the language of people
with mental illness:
It is often striking how well satisfied many of them are with their very inadequate
communication, showing little or no evidence of concern over its unintelligibility.
They either fail to recognize that you are having trouble or they are haughty
about your stupidity.
This comment does provide insight into the communication style some people with
schizophrenia used prior to the advent of anti-psychotic medication.
Cameron (1944) reported his work to the American Psychiatric Association in 1939.
Another researcher at the same conference reported on the speech production of
people with schizophrenia in a quasi-experimental study:
“...in the color-sorting test one of our patients picked out various shades of
green, but in doing so he named them as peacock green, emerald green, taupe
green, bright green, equet green, bell green, baby green. He refused to say that
all might be called green. Another patient said, in the same situation, “This is
the color of the grass in Virginia, this is the color of the grass in Kentucky, this is
the color of the bark of the tree, this is the color of the leaves” (Goldstein 1944,
p. 26).
Goldstein (1944) presents this as indicative of a type of failure in the communicative
task, that is, the ‘correct’ answer was in fact ‘green’, and the participants were unable,
Chapter 2 Language and mental illness
27
or perhaps unwilling, to produce this answer. Yet both respondents do provide answers
that suggest acute powers of observation even as they exceed the task at hand. The
second person’s reported choices bear this out even more spectacularly, evincing an
awareness and attendance to perceptual shifts that elude most of us. The second
response, provided under quasi-experimental conditions, actually falls into repetitive
and sophisticated rhythm, the dactylic, or ‘heroic’ hexameter used by Homer
(Thalmann 1984), thus demonstrating extraordinary skill at manipulating aspects of
language not restricted to simple conveyance of meaning.
Goldstein (1944, p. 25) is not insensitive to the motivation for this language choice,
despite it not meeting the ostensible goal of the test:
Analysis reveals that many of the very strange words which the patients use
become understandable when considered in relation to the concrete situation
which the patient experiences at the moment and which he wants to express in
words. In their language there is an absence of generic words which signify
categories or classes.
The link between these examples and metonymy is that what can be interpreted as
unrelated to the topic by a clinician or a researcher may be the result of the metonymic
selection of an attribute by the speaker that is not immediately clear to the listener
within the context. Here the determination of the context is set by the clinician or
researcher, hence the frustration in relation to the colour-sorting tests. For the
researcher, there is one ‘right’ answer, which is an accepted generalisation, conforming
to a specific level of categorisation. On the other hand, for the respondents the naming
task is correctly completed using terms sourced from their lived experience of colour,
illustrating Rosch’s (1978) observation that people frequently select category levels
based on embodied experience.
METONYMIC SPEECH IN SUBSEQUENT PSYCHIATRIC DISCOURSE
Metonymic distortion, subsequently referred to as metonymic speech, has attracted
variable interest as a concept in psychiatric research and clinical practice. Post-war
psychiatry was heavily influenced by the rise of behaviourism, and the introduction of
the first anti-psychotic medications. The term was included as a sub-type of the
category ‘word approximations’ in Andreasen’s Thought Language and Communication
Scale (1979a). It has resurfaced in a 2009 textbook, with one of the examples, of ‘a
menu’ standing for ‘a meal’ being taken (without credit) from Cameron (1944):
Chapter 2 Language and mental illness
28
Metonymy: Speech disturbance common in schizophrenia in which the affected
person uses a word or phrase that is related to the proper one but is not the one
ordinarily used; for example, the patient speaks of consuming a menu rather
than a meal or refers to losing the piece of string of the conversation rather than
the thread of the conversation. See also paraphasia and word approximation
(Sadock 2009, p. 925).
Another aspect of this reappearance is that metonymy is described as a ‘common’
speech disturbance in schizophrenia, despite minimal evidence to this effect
(Andreasen 1979b; Andreasen & Grove 1986).
The observations about metonymy and its influence on communication made by these
early researchers (Cameron 1944; Goldstein 1944) generate striking resemblances
with the way the concept is later mapped by cognitive linguists, but to date no
reference to this work in contemporary cognitive linguistics literature has been located.
STUDIES FROM THE SECOND HALF OF THE TWENTIETH CENTURY
While relatively neglected in psychiatric research in the second part of the twentieth
century, the language used by people with mental illness became a subject of interest
for researchers in other disciplines during this period. Chapman (1960), a linguist,
undertook an influential study of comprehension of literal and figurative language by
people with schizophrenia, people with brain damage, and a control group with no
known morbidity.2 Chapman looked at differences in how people with schizophrenia
figuratively misinterpret set examples of literal language, and literally misinterpret set
examples of figurative language, and identified a trend toward the latter, suggestive of
a certain fixity of the meaning of words held by people with schizophrenia that is
relatively insensitive to changes in context. Chapman (1960, p. 413) also expressed
frustration as a researcher:
Errors are made for many reasons besides the kind of misinterpretation under
investigation, and the bases of these errors are mostly unidentified. For
example, many schizophrenics are non-cooperative and respond carelessly,
mark randomly, mark by position, or mark in response to other characteristics of
the test which are difficult for the examiner to identify.
In fact, many of the researchers who decry the difficulty (or tedium) of communicating
with people with schizophrenia are in fact complaining of frustration at not receiving
2 The ‘control’ cohort for the study were noted to comprise 23 medical students from the
researcher’s own faculty, and 13 firemen stationed nearby.
Chapter 2 Language and mental illness
29
answers that suit their particular research or diagnostic question (Bleuler 1911/1950;
Chapman 1960; Goldstein 1944).
Goffman (1963, p. 129), a sociologist, highlighted the negative effects of
communication problems not just on the people who experience them, but also on the
broader community:
Failure to sustain the many minor norms important in the etiquette of face-toface communication can have a very pervasive effect upon the defaulter's
acceptability in social situations…A solution was for the individual who cannot
maintain an identity norm to alienate himself from the community which upholds
the norm, or refrain from developing an attachment to the community in the first
place. This is of course a costly solution both for society and for the individual,
even if it is one that occurs in small amounts all the time.
Goffman (1963) introduced consideration of what the experience of communication
problems is like for the individual. He also explicitly addressed the issue of the cost that
accrues to society as a whole as a result of this alienation of its members.
Rochester and Martin (1979) published Crazy talk: a study of the discourse of
schizophrenic speakers, a study focussed on speech production by people with
schizophrenia. They were critical of previous studies that automatically conflated
thought with speech. They were also critical of experimental linguistic studies,
preferring speech in context:
Investigators of schizophrenic speech have attempted to characterize those
features of the corpus that differ from normal. In effect, the effort has been to
describe the failures rather than the overall performance of the schizophrenic
speaker (Rochester & Martin 1979, p. 24).
They conducted a series of studies that attempted to bridge the divide between
experimental and context-based research, requesting participants in their cohort to
verbally describe pictorial cartoons they were shown, and also to re-tell brief narratives.
Their observations included that:
Taken most generally, the clinical and experimental evidence suggests that
schizophrenic patients attend more strongly to lexical items than to the contexts
in which those items are used. This means, roughly, that schizophrenic
speakers are more likely to focus on lexical meaning in their discourse than on
the meaning of whole clauses or the way in which clauses are related. And it
Chapter 2 Language and mental illness
30
indicates that schizophrenic listeners are more likely to be biased by the
"strong" or "preferred" meaning of a word than the meaning of a word in its
sentence or discourse context (Rochester & Martin 1979, p. 91).
This phenomenon can be linked to the position taken by cognitive linguists that
speakers generally hold an ‘encyclopaedic’ notion of the meaning of a particular word
rather than a ‘dictionary’ one (Croft & Cruse 2004).
Researchers who have investigated language for its capacity to assist with differential
diagnosis have conducted semi-structured interview schedules designed to avoid
discussion of pathology, in the interests of rendering the rater ‘blind’ to the diagnosis.
Interestingly, there was less attention to the influences of the so-called ‘neutral’ topics
chosen may have on the language choices, and hesitations of participants. For
instance, Andreasen & Grove (1986, p. 349) included discussion of ‘politics, religious
beliefs, and family life’. While such topics do not provide information about psychiatric
diagnoses, they stray into areas which have typically been considered to invoke social
awkwardness, which would then be a factor that would alter the data collected. Another
group used the Indiana Psychiatric Illness Interview, about which they say: ‘This
measure differs from other psychiatric interviews in that it does not introduce content’
(Lysaker et al. 2005, p. 67). However, by asking participants how their condition
‘controls’ their life and how they ‘control’ their condition, the researchers introduce very
specific content, inducing inferences of power that are leading in terms of language
choice.
At the turn of the new century, McKenna and Oh (2005) published a critical history of
language studies in schizophrenia, with particular focus on the ways in which
researchers have attempted to resolve the issue of the relation of speech to thought
disorder. They take up the concept of a breakdown in linguistic competence occurring
in schizophrenia:
Unlike patients with autism and Asperger's syndrome, who fail to develop the
ability to make inferences about the mental states of others, in schizophrenia
there would likely be a breakdown in an ability the patients once had. This
would cause them to continue to try and make inferences but to find the
process difficult and produce ones that were faulty (McKenna & Oh 2005, p.
118).
This echoes the abandoned concept of schizophrenia involving regression (Vygotsky
1934/1994), in that it describes a ‘breakdown in an ability the patients once had’.
Chapter 2 Language and mental illness
31
However, while the authors reported that people with schizophrenia may continue to
make failed attempts to correctly draw inferences, they do not suggest that they revert
to the kinds of inferences they made as children.
Approaches to language and mental health beyond
psychiatry
This section will consider how other disciplines approach the issue of language and
mental illness, and will consider those works that throw light on the current approach.
NURSING
Significant work that has been published within nursing literature that takes into
account the subtleties of communication and inference. Peplau, writing in the 1950s,
was strongly influenced by the psychodynamic work of the psychiatrist Harry Stack
Sullivan. She addressed her work to interpersonal relationships between nurses and
patients in all settings, including the role of communication within these relationships:
Nurses use words in many ways in their relations with patients. They
communicate facts, they converse about everyday events, they convey
interpretations of events that occur in carrying out the plan of treatment.
Concepts, words, or symbols chosen to express ideas, thoughts or feelings, or
to indicate objects referred to, often determine whether the tool -- language -will be helpful in the reshaping of experience or whether it will operate as a
detriment to elaboration of meanings of events (Peplau 1952/1988, p. 289).
The key element for the present study that was identified by Peplau is that the choices
nurses make in their communications with people can influence the outcome in terms
of the person’s experience. Peplau (1952/1988, p. 307) clearly states the approach to
language that nurses should adopt in approaching this work:
In relations with patients nurses can attend precisely to the way in which a
difficulty is stated by the patient. The exact wording is more important than
abbreviated or cryptic recordings of the complaint.
Sixty years later, Stein-Parbury (2014, p. 195) makes explicit the power that nurses can
exercise when they choose to pick up on inferences from the people they are caring
for, or ignore them:
Cues are small units of information that are part of a larger, more complex
phenomenon. They indicate a need for further exploration into the
Chapter 2 Language and mental illness
32
phenomenon...Sadly, nurses frequently either fail to acknowledge patient cues
or even actively discourage further exploration of them.
While these authors are not specifically addressing mental health nursing, their
observations apply to the field as well as to nurses in general settings.
CONVERSATIONAL THERAPY AND SYSTEMIC FUNCTIONAL
LINGUISTICS
There is a small body of work explicitly using the approach of Systemic Functional
Linguistics (SFL) (Halliday & Matthiessen 2004) to study therapeutic practice. SFL is a
highly-detailed descriptive method for analysing language processes, and therefore an
effective tool for describing the language people are using, and tracking language
change. Researchers have attempted to use tools developed in SFL to track change in
the language produced over time by people consulting therapists as a way of
measuring the effectiveness of that therapy (Meares et al. 2005). This collaborative
work between linguists and clinicians builds on his clinical use of figurative language,
particularly metaphor, within the practice of conversational therapy:
Language is an important determinant of which components of an affective
response will be differentially perceived and communicated in a particular
culture. For a language of affects to develop beyond the bodily, metaphor is
required (Meares 2005, p. 92).
Echoing Peplau’s (1952/1988) insistence on the importance of attending to the
person’s exact words, Meares (2005, p. 176) articulated the acuity practitioners need to
employ in listening in order to identify the key aspects of language that may be
developed for therapeutic effect:
By listening, in a particular way, to the exact words and the way that they are
used, one may begin to be able to use words that are repeated, words that
seem unusually loaded, or words that have a peculiar ring to them, as a means
of entering into previously unexplored areas of psychic life. A particular word or
phrase may contain within it a "micro-history" that can be fleshed out and
enlarged, and, in some cases, illuminate a much larger area of the individual's
ordinary existence.
Other researchers in the field have investigated whether observed changes in patterns
of language production correlate to actual changes in well-being or behaviour, but they
have not yet reached conclusive answers (Henderson-Brooks 2010; Muntigl 2004).
Chapter 2 Language and mental illness
33
Fine, a linguist working in the discipline SFL, published Language in psychiatry: a
handbook of clinical practice (2006) as a primer for clinicians interested in developing
their knowledge about language to improve their clinical practice. Rather than drawing
examples of language used in clinical practice empirically, Fine himself constructed a
short passage of text, and then imaginatively rewrote it in different ways, ostensibly to
demonstrate different language practices that may occur subsequent to different
mental health conditions. The resulting written constructions bear little resemblance to
the language clinicians actually encounter in their work.
PSYCHOLINGUISTICS AND NEUROLINGUISTICS
Observations about language and the mental processes that underpin its use in both
typical and atypical ways form the basis of modern experimental approaches.
Psycholinguistics is the subfield of linguistics whose goal is to discover the
psychological principles that underlie the ability of humans to comprehend,
produce and acquire language (Akmajian et al. 2010, p. 595).
Like Jakobson (1956/1987) these researchers frequently undertake research into
problematic use of language as a way to understand the fundamental processes of
language. The more recent sub-discipline of neurolinguistics narrows the field of study
further:
Neurolinguistics studies the relation of language and communication to different
aspects of brain function, in other words it tries to explore how the brain
understands and produces language and communication (Ahlsen 2006, p. 3).
Within these disciplines, researchers have examined both processing of figurative
language, and language and mental illness, with the former studies primarily
addressing metaphor, and the latter predominantly focused on schizophrenia.
Gibbs (2007c, p. 23) noticed the relative lack of attention to metonymy paid by
psycholinguists compared to metaphor, and asked:
Is there a reason for psycholinguists' failure to study conceptual metonymies in
figurative meaning construction, despite the prominence of work by cognitive
linguists such as Panther demonstrating the salience of metonymy in many
aspects of utterance interpretation? One difficulty in experimentally testing for
the presence of conceptual metonymies during utterance interpretation is that
these presumably entrenched knowledge structures are rather abstract.
Chapter 2 Language and mental illness
34
Current psycholinguistic research makes mention of the attention to metonymic speech
production in early psychiatric writing (Cameron 1944, Goldstein 1944), but this has
generally not translated into its incorporation into experimental designs, perhaps for the
reasons suggested by Gibbs (2007c).
A recent exception is the work of Rapp and colleagues (2008; 2011). They presented
the results of a study in which people diagnosed with schizophrenia and a control
group each had to interpret short sentences with literal, metonymic and nonsense
meanings, and determine if the meaning were acceptable or not. They reported that
people with schizophrenia made more errors in correctly attributing acceptability to
metonymic sentences (Rapp et al. 2008).3 They subsequently repeated the reading
exercise with a group of 14 people without any known history (including familial) of
mental illness, and monitored the responses using functional Magnetic Resonance
Imaging (fMRI) technology (Rapp et al. 2011). They reported that the region of the
brain most notably activated when subjects were interpreting metonymies is a region
associated with the integration of world knowledge with linguistic knowledge. They also
reported that subjects rated metonymically generated meanings as less acceptable
than literal ones, and that subjects demonstrated a great deal of variation in their
acceptance ratings for metonymic sentences. The authors reported a possible
explanation for this variation was the context-free single sentence paradigm they used
in their experiments. They report they plan to repeat the experiment with a group of
people diagnosed with schizophrenia, but to date, no results of this research have been
published.
Kuperberg, a key figure in the psycholinguistic study of schizophrenia, has published
extensively, producing both overviews and original research (Kuperberg 2010a, 2010b;
Kuperberg & Caplan 2003). Kuperberg (2010b, p. 590) posed the ‘What can
psycholinguistics bring to the study of schizophrenia…and vice versa?’ Her research is
focused on a technique involving the observation of Event Related Potentials (ERPs),
which show differential electrical activity in response to predicted and incongruous
words used in sample sentences (see, e.g. Kuperberg, Kreher & Ditman 2009). These
are investigated to, ’explore the relationship between semantic and syntactic
processing’ (Kuperberg 2010b, p. 593), picking up on the idea of semantic priming
initially suggested by Bleuler (1911/1950). They have demonstrated that some people
with schizophrenia exhibit different levels of electrical activity to controls in speech
comprehension exercises. The authors speculate that their research subjects have
This study has only been published as an abstract from the proceedings of the 1 st International
Schizophrenia Research Society Conference.
3
Chapter 2 Language and mental illness
35
problems in combining information from semantic memory with information in the
immediate context, specifically an overreliance on semantic association that can lead
to communication failures (Kuperberg 2010b). This work is providing experimental
evidence for theories put forward by earlier researchers (Bleuler 1911/1950; Chapman
1960; Rochester & Martin 1979).
Kuperberg (2010b, p. 594) also noted a key assumption that underpins the
neurolinguistic approach:
A related question is how an over-reliance on stored semantic associations
affects language production, which is how thought disorder is clinically
assessed and quantified. To date, the types of psycholinguistic paradigms used
to study schizophrenia have all measured language comprehension. The
underlying assumption has been that these comprehension studies tap into the
same semantic associative abnormalities that affect production. However, there
has been little work testing this assumption.
Another key problem this research faces is the gap between language as used under
experimental conditions, and language used in actual pragmatic exchanges. Kuperberg
and Caplan (2003, p. 451) note:
In one study using this paradigm, at least 70% of responses by both patients
and controls that were judged to be pathologic based on the word-association
test alone became meaningful in the context of sentences.
This demonstrates that an inference that had been made by a researcher that the
language used was pathological was not supported when the context for the language
use was taken into account. This echoes examples from earlier researchers (see e.g.
Chapman 1960; Goldstein 1944) where the context, or more specifically the
decontextualisation of language, influences the inferences that are based upon it.
A further limitation in this work is related to the narrow view of linguistics it adopts.
Kuperberg (2010a, p. 599) states that ‘traditional linguistic models have generally
regarded language as an insular system composed of its own sets of unique
representations and processes’, though this in fact describes one approach to
language, Chomsky’s Generative Grammar, which is a contested model (Harris 1993).
Moreover, Kuperberg (2010a, p. 599) reports that concurrent work in her laboratory:
Chapter 2 Language and mental illness
36
…suggests that the comprehension of simple visual events may engage some
of the same types of neurocognitive mechanisms as those which mediate
language comprehension.
This suggestion aligns with the view of language and its place in broader cognitive
processes posited by cognitive linguists, however, Kuperberg makes no mention of any
of the authors associated with this approach.
Another psycholinguistic approach to the study of language and mental illness draws
on the theory of mind. Originating in primatology, but soon developed in human
studies, theory of mind holds that in order to communicate effectively, particularly in
correctly interpreting inferences from another’s speech, a person must have the
capacity to recognise that the other person has their own mind and intentions (Brune
2005). Researchers have explored the idea that this theory of mind is impaired in
people with schizophrenia, using comprehension tests that require participants to read
short stories and correctly determine if characters in the stories hold false beliefs
(Champagne-Lavau & Stip 2010). They report a correlation between theory of mind
deficits and problems with pragmatic inferencing, which they suggest need further
investigation using fMRI techniques.
A third psycholinguistic approach involves the investigation of correlation between
speech disorder in people with schizophrenia, and specific failures in communication.
Docherty (2012) noted that, while it makes intuitive sense that neuropsychological
deficits and speech disorder are correlated, there is no robust evidence of specific
relationships. Docherty recorded interviews with people with schizophrenia who had
been asked to talk about themselves for ten minutes. Using a battery of analytic tools,
including the Thought, Language and Communication Scale (Andreasen 1979a),
Docherty (2012, p. 1333) reported her provisional conclusions, accompanied by a
caveat and a restatement of the intent behind this work:
Communication failures in the speech of schizophrenic patients may not reflect
psychotic ideation so much as they reflect combinations of schizophreniarelated neuropsychological impairments. That being said, the majority of the
variance in schizophrenic speech disorder is still unexplained. In the interests of
developing interventions to ameliorate speech disorder, it would be very helpful
to have a fuller understanding of its causes.
Other studies report similar constraints. In her summary of current psycholinguistic
research in schizophrenia Kuperberg (2010b, p. 595) states, ‘These ideas are fairly
Chapter 2 Language and mental illness
37
speculative’. The editors of a new journal devoted to cognitive research in
schizophrenia, also state that their field, like much neuroscience, reports provisional
knowledge of the relationship between the brain and the clinical manifestations of the
disorder (Green & Harvey 2014).
Reflecting the provisional results of much of this research, there have been few
attempts to translate findings into therapeutic interventions. This echoes the lack of
research into integrating language into cognitive remediation reported above in the
section, the Delivery of treatment.
NARRATIVE THERAPY
The work of the late Michael White and his colleagues is known as narrative therapy, or
narrative practice. The name and the approach originate in the metaphor that life can
be told as a narrative. This metaphor carries metonymic entailments, critically, that a
person can assume the role of ‘author’ of their lives, and select which aspects will be
given attention.
White cited select works from the early history of psychology (James 1890/1950;
Vygotsky & Kozulin 1934/1986), to inform his ideas, but also drew on a much broader
range of disciplines to develop his theory, including sociology (Goffman 1968), literary
theory (Bruner 1990) and anthropology (Geertz 1973). He also integrated the
philosophy of Foucault (1980), particularly his work on the operations of modern power,
into his approach:
We have a special responsibility to consider the ways in which we may have
unwittingly reproduced assumptions about life and identity that are disqualifying of
diversity in people's acts of living, and the ways in which we may have
inadvertently colluded with the power relations of local culture. Continually
questioning the metaphors we support in therapeutic conversations is part of this
special responsibility (White 2007, p. 31).
White clearly states the responsibility therapists have to consider their language
choices so as not to unwittingly reduce the recognition of the actions people already
undertake in their own lives. White frequently used metaphor in his therapeutic work
and his teaching. He spoke of providing ‘scaffolding’ to support people safely while they
engaged in rebuilding their lives. He also spoke of ensuring people had an ‘island of
safety’ before they embarked on difficult encounters, and expanded the entailments of
the phrase, through ‘peninsula’ to ‘continents of safety’. He responded to criticisms of
people who challenged his use of combat metaphors, stating that while he never
Chapter 2 Language and mental illness
38
introduced these particular metaphors into therapeutic conversations, he respected the
choices of people who viewed themselves as ‘fighting for their lives’ to employ such
terms, and would work with these metaphors with them for as long as they were useful.
He challenged metaphors deeply embedded in Western culture, such as the idea of
‘human nature’ as fixed, immutable and explanatory, and identified the typical
consequence in terms of the opportunities that are available for people who are
experiencing problems:
When human action is assumed to be a manifestation of some element or
essence of a self that is determined by human nature, or by a distortion of
human nature, it is rare for people to be invited to reflect on their lives in a way
that allows them to determine what certain events might say about what is
important to them (White 2007, p. 53).
White (2007) utilised the metaphor that life can be told as a story to create spaces
where people can reflect on their lives. In doing so, he capitalised on the ways in which
stories are not fixed, drawing on earlier theories of narrative:
The world is full of partial stories that run parallel to one another, beginning and
ending at odd times. They mutually interlace and interfere at points, but we
cannot unify them completely in our minds. In following your life-history, I must
temporarily turn my attention from my own. Even a biographer of twins would
have to press them alternately upon his reader's attention (James 1907/1987, p.
548).
White (2007) spoke of people presenting with ‘problem-soaked narratives’, in which the
problem they were dealing with had become completely entwined with their sense of
self or identity, sometimes with the collusion of others, including family members and
mental health workers. A characteristic of these narratives was that they would include
any and all information that contributed to the sense of the person themselves as the
problem, and at the same time, ‘edit out’ information which did not support this view.
Starting from the phrase ‘unique outcomes’ borrowed from Goffman (1968), White
(2007) mapped out a process for helping people to separate from totalising identity
descriptions. Called externalising conversations, he would start by encouraging a
person to think of an instance when they didn’t wholly embody the problem position.
From this point, he would ask them to expand on elements that contributed to the
‘unique outcome’, and slowly build up a picture of the person existing outside the
Chapter 2 Language and mental illness
39
problem definition they had occupied. The first step in these conversations is,
‘Negotiating a particular, experience-near definition of the problem’ (White 2007, p. 40).
The first word is ‘negotiation’, which signifies that the therapist supports the person to
develop such a definition of the problem. ‘Experience-near definition’ encourages the
person to describe what actually happens, rather than using general, underspecific
terms. In cognitive linguistic terms, this is language that does not function through
indirect reference, for example, metonymy. He states the intended outcome of this
approach:
In the context of externalising conversations, the problem ceases to represent
the “truth” about people’s identities, and options for successful problem
resolution suddenly become visible and accessible (White 2007, p. 9).
The links between narrative therapy and metonymy will be addressed in more detail
within Chapter 3: Metonymy.
Language and identity as experienced by people who live
with mental illness
What is often elided in the research history on language and mental illness is
information about the experience of language by people who live with mental illness.
There is a small body of knowledge about what it is like to be labelled with a mental
health diagnosis. This comes from the literature of first-person accounts (see, e.g.
Barnes & Berke 1971).
Barham and Hayward (1991) published a book length study From the mental patient to
the person based on semi-structured interviews with 24 people with schizophrenia who
were living in the community after experiencing lengthy periods in psychiatric hospitals.
The book includes excerpts from the interviews, and the analysis highlights a number
of themes that are relevant to this study, including the reconstruction of personal
identity after psychosis, the experience of stigma, and positive and negative
encounters with mental health workers. Much of this is reported in the participants’ own
words, and the authors are careful not to erase contradictions, dilute harsh appraisals
or edit out the humour in their accounts. In their discussion, the authors describe the
two constructions of identity that people who live with mental illness in the community
contend with, one in which their personhood is foregrounded, and the other in which
their status as mental patient is placed as the salient attribute:
Chapter 2 Language and mental illness
40
Though bracketed, the category ‘mental patient’ is none the less still active and
the P (mp) [Person (mental patient)] predicament may be seen to identify the
structurally and culturally unresolved character of the terms of membership and
participation available to people with mental illness in social life.
The relevant contrast is with what we may characterise as the Mental Patient
(person) predicament of MP (p) in which the person comes to be defined in his
or her illness and the category of the person is bracketed but not necessarily
abolished (Barham & Hayward 1991, p. 144).
They suggest that the P (mp) predicament potentially offers a means to move beyond
being identified primarily through the experience of mental illness, but note, ‘from the
testimonies of our participants…it is clear that the accent here is very much on the
potential’ (Barham & Hayward 1991, p. 144). The conceptualisation, and the notation in
which they express it, illustrate the operation of metonymy, with inferences following
from which element, the person or the illness, is foregrounded.
More recently, the Scottish Recovery Network has collected a significant number of
narratives of recovery from mental illness, and published them on the internet. These
stories have use in and of themselves, as reading other people’s descriptions of their
experiences has been reported as helpful. A selection from the narratives has been
subjected to analysis, with a focus on identifying which processes people with mental
illness engage with support them in their recovery (Brown & Kandirikirira 2007).
Metonymic language is not explicitly mentioned, but its influence is evident in the
descriptions of identity; for some participants, forming identity based on having a
mental health diagnosis was reported as helpful in giving them a sense of community
and connection when dealing with an otherwise isolating experience.
The mental health consumer movement is creating a space where people can talk
about their experience, frequently in online social media in the form of webpages, blogs
and twitter accounts. Some of these include robust discussion of language use, for
instance, critiques of language used about people who live with mental illness that
appear in the media or in policy documents (see, e.g. Epstein n.d.; Webb n.d.).
One of the most prominent Australians with lived experience of mental illness is Janet
Meagher, who recently completed a term as one of the inaugural National Mental
Health Commissioners. Meagher (2014, p. 8) firmly positions the significance of her
experience of mental illness within her overall sense of self:
Chapter 2 Language and mental illness
41
I have accepted schizophrenia and no longer deal with it as a problem needing
to be ‘treated’ or eliminated, but as a ‘normal’ aspect of me that merely needs to
be managed and monitored by me. This mindset is hard to practise and learn
but makes for a great quality of life and the ability to reach some of my
remaining potential.
Meagher’s (2014) comment, that dealing with schizophrenia as a normal part of oneself
is hard, reflects clinical and research discourses that address the condition through the
lens of impairment. Her assessment of the contribution of this practice to her quality of
life suggests the benefits of applying rigorous method to an alternate view, developed
through her lived experience and that of others who live with mental illness.
STIGMA
The experience of stigma, frequently expressed through language, is widely reported
by people who live with mental illness (Brown & Kandirikirira 2007; National Mental
Health Commission 2012; Reavley & Jorm 2011). Stigmatising attitudes voiced and
enacted by clinicians have been reported toward people who live with mental illness
(Hill 2010; Ross & Goldner 2009), and these have been associated with poorer
outcomes through the process termed diagnostic overshadowing, where physical
symptoms are either not recognised, or incorrectly attributed to mental illness (Shefer
et al. 2014). Self-stigma, where stigmatising attitudes are internalised by people who
live with mental illness themselves, has been associated with having greater insight
into the illness (Lysaker, Roe & Yanos 2006), and is also linked to poorer outcomes
(Corrigan, Kosyluk & Rusch 2013).
How stigma functions pragmatically has also been addressed by Thornicroft and
colleagues (2007, p. 192), who state that:
Stigma can therefore be seen as an overarching term that contains three
elements: problems of knowledge (ignorance), problems of attitudes (prejudice),
and problems of behaviour (discrimination).
They call for a shift in focus from attending to attitudes, for example, through population
surveys (e.g. Reavley & Jorm 2011), to evidence of behaviour, for example, how many
people with disability does your company actually employ?
The influence of metonymy on the practice and experience of stigma is considered in
more detail below in Chapter 3: Metonymy, in the section ‘Labelling theory and stigma’.
Chapter 2 Language and mental illness
42
Conclusion
The critical role played by language in the assessment of and therapeutic approaches
to mental illness has been a subject of research from the beginning of modern
psychiatry. Clinical decisions that are made, based on assessment of a person’s use of
language, carry considerable diagnostic and medico-legal implications. Yet both
research and clinical practice have frequently been characterised by received and
conventional views about language that do not reflect the complexities of language and
communication explored within linguistics. Conversely, therapeutic approaches that do
focus on careful use of language in order to create opportunities for people to
participate in their own recovery (e.g. Meares 2005; White 2007), do not demonstrate
attention to the specific illness-related communication problems that people with mental
illness may experience. ‘Illness narratives’ have traditionally run the risk of being
dismissed as merely anecdotal and peripheral to the type of research that should
inform health practice (Bury 2001). Blogs are considered to be unreliable sources for
the purpose of research.
Previous studies have considered unusual patterns of language use as a tool to assist
with differential diagnosis of mental illness in the clinical setting (Andreasen 1979a;
Chapman 1960). Contemporary psycholinguistics and neurolinguistics elicit examples
of language use under experimental conditions in order to locate areas of the brain
involved in language processing, and speculate on aetiological causes of language
dysfunction (Crow 2010; Kuperberg & Caplan 2003). First person accounts of mental
illness broadly discuss what it is like to live with mental illness, recounting clinical and
social events that have shaped their experience (Barnes & Berke 1971; Brown &
Kandirikirira 2007).
This study is designed to reflect the complexity of the topic, as it is expressed through
the language of people living with mental illness. The next chapter gives a more
detailed background to one aspect of language, metonymy.
Chapter 2 Language and mental illness
43
Chapter 2 Language and mental illness
44
CHAPTER 3: METONYMY
This chapter is a review of metonymy as it has been discussed, initially within rhetoric,
and more recently in cognitive linguistics. At a basic pragmatic level, metonymy is a
case whereby a speaker says one thing, but means another. As a communication
strategy, a speaker using metonymic language relies on confidence that, in the context,
the hearer will be able to infer the intended meaning rather than the literal one.
Earlier definitions
The word metonymy originates from the Greek ‘metonymia’ from ‘meta’, after or
beyond, and ‘onymia’ a suffix used to name figures of speech, from ‘onoma’, name.
This is usually abbreviated to meaning ‘change of name’. As such, it is not specific, as
the same definition could be used for other figures of speech, including metaphor. It
was considered an aspect of rhetoric, or the art of persuasion using words, and
therefore treated with some suspicion; an educated person should be aware of it in
order not to be manipulated in their thinking (Ricoeur 1975/2003).
Further refinement of this figure of speech was taken up by Latin rhetoricians:
Quintilian lists the kinds usually named: container for thing contained (“I’ll have
a glass”); agent for act, product, or object possessed (“reading Wordsworth”);
cause for effect; time or place for their characteristics (“a bloody decade”);
associated object for its possessor or user (“the crown” for the king) (Preminger,
Warnke & Hardison 1986, parentheses in original).
Within rhetoric, figures of speech were considered to contribute to the style of one’s
language, and able to be modified to achieve different ends. As rhetoric declined,
tropes, or figures of speech, were relegated to poetics, and style treated as
ornamentation for literary purposes. Ricoeur (1975/2003, p. 10) comments that, ‘Before
becoming futile, rhetoric was dangerous’. It is ironic that the one discourse in which
metonymy is still frequently and deliberately used is advertising, in which
communication is explicitly designed to persuade people to make choices they may not
otherwise make.
It is the traditional definition of metonymy, as a figure of referential substitution, that
Cameron (1944) and Goldstein (1944) used in describing how metonymic speech
featured in their clinical population. However, rather than a deliberate linguistic strategy
Chapter 3 Metonymy
45
used to influence communication, they posit its use as pathological and obscuring
meaning.
JAKOBSON
Jakobson, a Russian linguist, published an article, Two aspects of language and two
types of aphasic disturbance (1956/1987), which offered a radical reinterpretation of
the relation of metonymy and metaphor. Jakobson describes language as consisting of
two processes, the selection of terms from the range of possibilities and their
combination into phrases and sentences. Jakobson observed two distinct patterns of
dysfunction in the speech production of people with aphasia, difficulty in selecting
terms and difficulty in combining them. From these observations, Jakobson posited two
poles of speech, the metaphoric and the metonymic poles. In Jakobson’s view,
metaphor relates to the process of selecting a reference for something, a relatively
unconstrained process. Metonymy relates to the combination of selected terms into
coherent structures, and is therefore more subject to constraint. Though he did not
undertake empirical research into this phenomenon Jakobson (1956/1987, p. 110)
stated that:
In normal verbal behaviour both processes are continually operative, but careful
observation will reveal that under the influence of a cultural pattern, personality,
and verbal style, preference is given to one of the two processes over the other.
Jakobson’s (1956/1987) radical departure from traditional views of the function of each
of these figures of speech, metaphor and metonymy, has been both influential and
controversial. The elevation of metonymy to a standing equivalent to metaphor, and the
suggestion that ‘in normal verbal behaviour both processes are continually operative’
prefigure the cognitive linguistic approach of viewing these figures not just as
ornaments, but as part of ‘everyday speech’. The association of metonymy with the
combinational aspect of language has largely been abandoned in subsequent linguistic
research, but was taken up by French structuralists, in interpreting topics in disciplines
such as semiotics (see e.g. Barthes 1994) and psychoanalysis, in which Lacan (2006)
suggested that the unconscious is structured like a language. Jakobson’s (1956/1987)
ideas about the potential function of metonymy, particularly in everyday language,
would be most fully explored within cognitive linguistics.
Chapter 3 Metonymy
46
Cognitive linguistic definitions
Cognitive linguistics is an approach that considers language as one cognitive process
among other cognitive processes, including attention, problem-solving, and
categorisation (Geeraerts & Cuyckens 2007). By directly addressing cognitive and
conceptual processes, cognitive linguistics is ideally suited to considering language as
it operates in psychological states. That said, it has been minimally used for this
purpose. One of the core principles of cognitive linguistics was noted to be that
linguistic meaning is encyclopaedic and non-autonomous (Geeraerts 2008). Langacker
(2007, p. 432) expands on the pragmatic implications of this concept:
On the encyclopedic view, a lexical item provides a particular way of accessing
associated domains of knowledge. The access it affords is flexible and subject
to contextual influence, but not at all random or unconstrained.
It is this capacity to activate one out of a range of possible meanings that allows
listeners to infer the correct meaning intended by the speaker, by using the context of
the speech event, the preceding language (sometimes specified as the cotext) and
their real world knowledge. A key element in this process is the role played by
constraint, without which an intended meaning could not be conveyed.
Within cognitive linguistics, attention to metonymy has developed as a key focal area
for the discipline. Being a subject of attention in a burgeoning discipline does not
guarantee agreement among scholars. Defining metonymy in cognitive linguistic terms
remains controversial (see e.g. Benczes, Barcelona & Ruiz de Menoza Ibanez 2011;
Croft 2006; Peirsman & Geeraerts 2006a, 2006b). This demonstrates the prescience of
earlier commentators:
In seeking a logic underlying metonymy and other tropes, theorists are forced to
redefine them, reassigning some of the examples they traditionally include to
other tropes, and discarding others. In so doing, they define the tropes
stipulatively, or figuratively (Preminger, Warnke & Hardison 1986).
An early formulation comes from the book recognised by many as the seminal text in
cognitive linguistics, Metaphors We Live By:
In these cases, as in the other cases of metonymy, one entity is being used to
refer to another...But metonymy is not merely a referential device. It also serves
the function of providing understanding (Lakoff & Johnson 1980, p. 36).
Chapter 3 Metonymy
47
This recalls the etymological definition of metonymy in its under-specificity, and again
fails to distinguish metonymy from metaphor and other figures of speech. As the title
suggests, the authors are not yet ready to grant metonymy equivalent attention as
metaphor. Nonetheless, it is a statement of a role for metonymy in language, including
everyday speech, which takes it beyond simple ideas of referential substitution.
The difference between metonymy and metaphor is a topic of longstanding
controversy. One proposal is that in metaphor, the source and target are linked by
similarity, whereas in metonymy, they are linked by association. A simple test has been
used to differentiate the figures: in metaphor, the source ‘is like’ the target, in
metonymy, the source ‘stands for’ the target. For instance, it can be said that, ‘a man is
like a lion’, but not, ‘a man stands for a lion’, therefore this is a metaphor. Or it can be
said, ‘the Crown stands for the monarchy’, but not, ‘the Crown is like the monarchy’,
and therefore this is a metonym. Cognitive linguists have suggested that in metaphor,
the source and the target exist in different conceptual domains, but in metonymy, they
exist in the same conceptual domain. The distinction remains controversial (Croft 2006;
Peirsman & Geeraerts 2006a, 2006b).
Lakoff studied metonymy further in his book Women, Fire and Dangerous Things.
Following on from Rosch’s work on categorisation (Rosch 1978), he explored ways in
which categories could include members which stand as prototypes, as well as more
peripheral members:
A cognitive model may function to allow a salient example to stand
metonymically for a whole category (Lakoff 1987, p. 90).
This metonymic process can have significant influence, as when the category as a
whole is invoked, the salient prototype occupies the foreground, relegating peripheral
members to the background.
Cognitive linguistic work on metonymy accelerated in the 1990s, and a definition
formulated toward the end of the decade has been widely used since then:
Metonymy is a cognitive process in which one conceptual entity, the vehicle,
provides mental access to another conceptual entity, the target, within the same
idealized cognitive model (Radden & Kovecses 1999, p. 21).
However, there are a number of problems with this definition. It relies on a specific
concept unique to cognitive linguistics, the idealised cognitive model, or ICM (Lakoff
Chapter 3 Metonymy
48
1987). Lakoff proposes that ‘we organize our knowledge by means of structures called
idealized cognitive models’ (1987, p. 68), and that:
Each ICM is a complex structure whole, a gestalt, which uses four kinds of
structuring principles: propositional structure…image-schematic
structure…metaphoric mappings…[and] metonymic mappings (1987, p. 68).
Littlemore (2015, p. 12) expands on the role and development of ICMs:
The best way to describe them is as a series of embodied, encyclopaedic,
loosely connected and somewhat idiosyncratic knowledge networks that we
have in our minds. But how do they get there? The explanation usually given is
that ICMs build up gradually over time as a result of our interactions with the
world and the people in the world. In other words, they are largely ‘usagebased’.
While influential, the term is not agreed to by all within cognitive linguistics, and is not
broadly recognised outside the field. Critics have focused on problems with defining the
borders of an ICM, which renders a definition reliant on the process occurring ‘within’
this border somewhat tenuous (Feyaerts 2000).
An important aspect of the cognitive linguistic approach to metonymy is that it offers
insights into the cognitive motivation for using metonymy in the first place. These
centre round the ideas of economy and optimalisation:
In order to identify a certain referent, make use of a feature that economically
allows for the unique identification of that referent in the given context
(Geeraerts & Peirsman 2011, p. 96).
Economy of language use was most famously addressed by H. Paul Grice, a
philosopher of language who has had a major influence on the development of the
pragmatic approach to linguistics. Grice (1975, p. 45) stated a number of maxims that
govern how conversation is understood, including the maxim of quantity:
Make your contribution as informative as is required (for the current purposes of the
exchange),
Do not make your contribution more informative than is required.
When the ‘current purposes of the exchange’ are well-known to both speaker and
hearer, short cuts can be taken that fulfil the second part of the maxim, that is, do not
provide more information than is required, while still comprising effective
Chapter 3 Metonymy
49
communication. Metonymic usage contributes to this economy. A term may appear
underspecific to an outsider, but function pragmatically in context.
Langacker (2009, pp. 44-5) addresses the discrepancy between source and target
(here called profile/active zone), and pragmatically inferred meaning:
We easily manage the imprecision and indeterminacy inherent in profile/active
zone discrepancy. In fact, we do not even notice it. The reason is that we are
able to make sense of discrepant expressions by exploiting general
knowledge…But why does profile/active zone discrepancy occur in the first
place? It is actually both natural and often necessary from the cognitive
standpoint. In many cases complete precision and accuracy in describing a
relational participant is simply not possible. Discrepant expressions are natural
because they profile, and thus make linguistically prominent, entities that have
greater cognitive salience.
Key elements to note are that indeterminacy frequently goes unnoticed, and that
complete accuracy in expression is neither possible nor necessarily desirable. At the
same time, it is the combination of indeterminacy, and the fact that indeterminacy can
go unnoticed that can lead to problems in the influence of language in mental health
practice, especially when the lack of noticing occurs in contexts of power imbalance.
A WORKING DEFINITION
The definition of metonymy with the best fit for this project comes from a recent
collection, Defining Metonymy in Cognitive Linguistics: Towards a consensus view
(Barcelona 2011). The book is a collection of papers from a conference that shared the
aim made explicit in the title. The book itself demonstrates a lack of consensus, which
early critics have noted, nonetheless, it yields at least one definition that solves some
of the earlier problems:
Metonymy is an asymmetric mapping of a conceptual domain, the source, onto
another domain, the target. Source and target are in the same functional
domain and are linked by a pragmatic function, so that the target is mentally
activated (Barcelona 2011, p. 52).
In this definition, the key concepts of target and source, mapping, asymmetry,
functional domain, pragmatic function and activation are useful in explaining
metonymy, especially in relation to its differentiation from metaphor.
Chapter 3 Metonymy
50
TARGET AND SOURCE4
The target is the intended meaning of an utterance. The source is the term selected to
suggest this target. Thus, when a nurse refers to ‘bed 15’, the intention is that a
colleague will understand they are referring to the person currently occupying bed 15;
the target is the person, and the source is their bed number. This raises a very
important aspect of metonymy: its dependence on contingency. In this example, the
person (the target) can only be effectively referred to by this bed number (the source)
while they are in it. Should they leave hospital, it will be meaningless to call them bed
15. Should they move to another bed in the unit, and somebody else occupy bed 15,
that new incumbent is the only one who can be effectively referred to in this way,
indeed it would be incorrect, perhaps catastrophically, to retain the initial source to refer
to the previous target. Authors have associated the contingent aspect of metonymy
with its use being accidental, as opposed to necessary (De Man 1979), or defeasible
(Panther & Thornburg 2007).
Another critical factor about the relationship between source and target for this study is
that when a source term is used, even if the intended target is successfully understood
by the listener in the context, elements of the source meaning are also mentally
activated. This plays a key role in the creation of inference, or the more subtle
nuancing of a message. For example, when a nurse refers to a person as ‘bed 15’,
their colleague may immediately and correctly understand which person is meant, but a
message that all of the people who occupy the beds are interchangeable has also been
made available. Brdar-Szabo and Brdar (2011, p. 236) have referred to this as, ‘an
efficient way of saying two things for the price of one’. It is possible for these secondary
meanings to be intentional, correctly inferred, welcomed or disavowed by different
speakers and listeners, and this is a key aspect in metonymy’s discursive power.
MAPPING
Mapping itself relies on a metaphor, that three-dimensional space can be represented
two-dimensionally, and that there are conventions to this representation. The purpose
of a map is to enable communication of information in economic means, to serve a
specific function. If a traveller can make their way safely through a landscape without
having to carry a three-dimensional model, their passage will be more efficient. Using
this metaphor to describe communication, the fewer words one needs to be understood
in a specific context, the more efficient the communication will be. For the map user, as
4 Other terms that have been used to denote the Target/Source dyad are Tenor/Vehicle
(Richards 1991), trajector/landmark (Langacker 1993), and profile/active zone (Langacker
2009).
Chapter 3 Metonymy
51
for speakers and hearers, understanding of, and concordance with the conventions is
necessary for the process to work.
Mapping also invokes traditional ‘common-sense’ views of language, most particularly
that there are things and concepts in the world, and that words name these, in a
straightforward manner. It is this view which prevails among many language-users, but
that has been comprehensively refuted by linguists.
Mappings become culturally and lexically entrenched, and as Turner (1991)
shows, they actually define the category structure for the language and culture.
Rather remarkably, although the vocabulary often makes the mapping
transparent, we are typically not conscious of the mapping during use, and in
fact are liable to be surprised and amused when it is pointed out to us
(Fauconnier 1997, p. 9).
This observation states another aspect of the concept that language choices may be
conscious or unconscious.
ASYMMETRY
The reference to asymmetry denotes that in metonymy, it is usually one attribute that is
selected for highlighting. Another term used to describe this is ‘foregrounding’. It is this
selection of an attribute for highlighting that functions pragmatically to suggest what
inference is to be understood.
Asymmetry contributes to distinguishing metonymy from metaphor, the latter being
considered to map symmetrically. When a concept is mapped metaphorically,
conceptual entailments tend to follow. For example, in the metaphor ‘love is a journey’,
a whole array of ideas can be transferred, people can be understood to be at a
crossroads, travelling hard, on the rocks, or in for a bumpy ride (Lakoff 1987). When
metonymy is used, this does not occur automatically, though some authors have
pointed out that metonymies can be linked in chains, to support coherence and build
inferences (Brdar-Szabo & Brdar 2011).
FUNCTIONAL DOMAIN
Functional domain replaces ICM from the earlier definition (Radden & Kovecses 1999),
and is more readily understood outside cognitive linguistics. A functional domain
describes the context in which speech occurs. A familiar functional domain is a hospital
ward. There is a broadly understood function in each specific ward, and this provides a
baseline context. Certain aspects of a functional domain may be transferable across
Chapter 3 Metonymy
52
units, for example, referring to people by their bed numbers, but others may not be
recognisable one floor up, for example, ‘CABG’ (Coronary Artery Bypass Graft,
pronounced ‘cabbage’ in cardiac inpatient units), which is only going to be familiar to
specialist cardiac teams. Slippage in metonymic referencing explains the ongoing and
unresolved notion of what to call a person who receives health care in different
settings. Having left hospital, a person may no longer consider themselves to be a
patient, with its connotations of passivity. A doctor may unconsciously wish to
foreground their role as expert in continuing to refer to the person as their patient.
Many opt for consumer, as connoting the person’s consumption of service, and this
connotes choice, even when the person may have no choice at all about receiving
service, for example, when they are legally constrained to accept treatment.
PRAGMATIC FUNCTION
Pragmatic function refers to how language actually works. It is less about the rules of
language, covered in semantics and grammar, than how language is effective in real
communication. For the purpose of this study, the pragmatic elements of particular
importance involve inferencing, that is, how the speaker’s intended meaning is
understood by the listener in the specific context. This can involve correctly interpreting
an indirect reference (metonymy), or presuming information not explicitly stated
(ellipsis).
ACTIVATION
The final part of the definition to note is the idea of activation. This is more than simply
referring. Words can carry a number of meanings, and language is processed ‘online’,
having been primed by the context. A classic example is ‘bank’: the sentence, ‘I have to
get to the bank before five’, will most readily be understood as referring to a financial
institution. However, if you are in a boat, you will understand the term is referring to the
bank of a river, without first having to process and reject the more common use of the
term. This process depends on context to determine which particular meaning is salient
(Giora 2003). Thus, use of the bed number on a ward will activate the concept of the
person occupying the bed. Of course it is possible that it is the bed itself that is being
directly referred to, for example, when someone is being enjoined to clean it, but again,
context will constrain which aspect is activated. Psycholinguistic research has
confirmed that native speakers of a language need no more time to process figurative
language than literal language (Gibbs & Colston 2006).
Chapter 3 Metonymy
53
Activation relates to asymmetric mapping and inference:
Metonymy-driven re-conceptualization can be relatively transient and limited...
or it can leave an extensive permanent trace in cognitive domains and in
inference patterns (Barcelona 2011, p. 13).
This concept is particularly pertinent in health settings. Some uses of metonymy can be
completely contingent, for example, referral to person by bed number they happen to
occupy on a particular day in a particular ward, and this will have only transient
meaning. However, in health discourse, attribution based on diagnosis can leave ‘an
extensive permanent trace’ (Barcelona 2011, p. 13).
Types of metonymy
Different authors have identified different subtypes of metonymy, based on the part of
speech affected, and the pragmatic speech function of the process. Three subtypes are
reviewed here as relevant to the thesis topic; referential, predicational and illocutionary
metonymy.
REFERENTIAL METONYMY
Referential metonymy is the prototypical form of metonymy, in which indirect reference
is made to an object, entity or concept, through substitution of the name of something
which is conceptually contiguous to that object. Frequently encountered referential
metonymies included the substitution of a part for a whole, or a container for its
contents.
PREDICATIONAL METONYMY
Predicational metonymy occurs when the transfer of meaning is in relation to properties
of the object. The term is based in grammar, where a sentence is divided into a subject
and a predicate, which carries information about the subject.5
Panther and Thornburg (2003a, p. 4) include predicational metonymy in their typology
of the figure, and illustrate it as follows:
An example of a predicational metonymy is ‘General Motors had to stop
production’, where the necessity or obligation to stop production stands for the
actually occurring event of stopping production (Obligation to act for action).
The metonymy involved is an instance of a high-level metonymic principle that
Care must be taken not to confuse the conceptual use of the word ‘object’ in linguistics, and
the grammatical use of the word, in which the ‘object’ is differentiated from the grammatical
subject.
5
Chapter 3 Metonymy
54
is very common in English and other languages. A potential event (e.g. the
ability, possibility, permission, obligation to undertake an action) is
metonymically linked to its actual occurrence.
The difference between referential and predicational metonymy is most clearly
described by Nunberg in his article ‘Transfers of Meaning’ (1995). He uses two
examples of sentences a person could say when handing keys to a valet:
1. This is parked out back.
2. I am parked out back (Nunberg 1995, p. 110).
Both of these sentences involve metonymy. In example 1. ‘This’ indirectly refers to the
car, which is not present, and the object which ‘This’ directly refers to is the key, which
is linked to the car by contiguity, that is, the key is needed to start the car. The
substitution that has occurred is in the reference to the object. In example 2. However,
the person speaking does not stand for the car. The substitution that has occurred is at
the level of the predicate, which describes a property of that person, in this instance,
that their car is parked out the back. Again, there is a metonymic transfer that is salient
in the context (handing keys to a valet), but it is at the predicational, rather than the
referential level.
Predicational metonymy can be used as a short-cut to convey meaning, when attention
is drawn to a property or attribute of an entity that it cannot literally possess, but which
is readily understood. For instance, the University of Technology adopts the term
‘Ambulant toilet’, for which a literal reading is impossible, to signpost a toilet accessible
to people who have difficulty ambulating. A prototypical examples of this type of
metonymy is ‘healthy diet’, in which health, is posited as an attribute, rather than a
result of a diet (Littlemore 2015). This technique is frequently used in advertising,
which extends the metonymic links in terms such as ‘healthy floor’, in order to sell floor
cleaners.
ILLOCUTIONARY METONYMY
Illocutionary metonymy is a form that demonstrates the intersection of cognitive
linguistics with pragmatics, in particular speech act theory. Panther and Thornburg
(2003a, p. 4) state:
The basic idea is that an attribute of a speech act can stand for the speech act
itself in the same way that an attribute of a person can stand for the person.
Chapter 3 Metonymy
55
Consider the requests, ‘Can you close the window?’ or ‘I would like you to close the
window’. These are examples of indirect requests, directed to another person, to close
a window. But both carry information in addition to the direct request, which could
simply be stated as ‘Please close the window’. The first also incorporates a
predicational metonymy, where the person’s capacity to perform the act stands for
them performing the act. The second includes information about how the act will affect
the person requesting it. In both cases, the indirect request stands for the direct
request. These are typical examples of the language practice known as hedging
(Eggins & Slade 1997), where a person does not directly express their meaning,
usually for external reasons. The process can extend further, for example, when a
person further away from a window asks, ‘Don’t you think it’s cold in here?’ couching
the request to close the window in even more indirect fashion.
Illocutionary metonymy creates space for a range of inferences, and subsequent
responses, for example, a literal response, ‘yes’ or ‘no’ to the last question. These can
be used for humorous effect. A person can follow someone else’s statement ascribing
intent where it is unlikely. For example, when someone declines a second serve, ‘Don’t
you like my cooking?’
There are a number of key elements from this discussion of metonymy from a cognitive
linguistic perspective that are relevant to the current study. These include:
1. Metonymy is prevalent in all speech contexts, including everyday speech, and
clinical encounters.
2. Metonymy is a conceptual process, not simply a linguistic substitution process.
3. Metonymy is valued because it allows for effective communication with efficient
cognitive work.
4. The choice of source to activate target may influence the pragmatic inference
that follows. This influence may be intentional or not.
These principles can be demonstrated through a consideration of how metonymy
operates in nursing contexts.
Chapter 3 Metonymy
56
Uses of metonymy in nursing contexts
Following Nunberg (1995), it is possible to see metonymy operating at the referential,
predicational and illocutionary levels in typical mental health nursing language, even
occurring concurrently. Consider the following examples:
1. My borderline needs some PRN.6 (Nurse to colleague)
2. That borderline needs some PRN. (Nurse to colleague)
3. I need some PRN. (dispensed for my borderline) (Nurse to colleague)
4. I need some PRN. (charted for my borderline) (Nurse to doctor)
5. I need some PRN. (Nurse to colleague)
The first two cases demonstrate referential metonymy – both use attributes of a person
that are salient in the ward context to refer to a person, in this case a diagnosis.
Example one can also function as an illocutionary metonymy; dispensing PRN
frequently necessitates a second nurse to countersign the medication chart, and so the
statement made to a colleague can also stand for a request for the colleague to
perform the requested action. Example 2 can also carry a range of pragmatic
inferences. It may be a statement of clinical opinion, which again includes a presumed
action in response. However, it may also be stated as an opinion by someone not
involved in the person’s direct care, and carry an implicit negative inference about the
care being delivered by the treating team.
Examples 3 and 4 demonstrate predicational metonymy, (note that I have also spelled
out in parentheses the indirect object which would typically be elided in these
exchanges). The nurse does not literally need the medication for themselves, so the
transfer of meaning does not occur at the level of the subject, but rather at the
predicate level, relating to a property of the indirect (and typically elided) object of the
sentence, the person for whom the PRN is intended. Both of these sentences are likely
to be uttered as written, but in fact, each has a different intent. Example 3 may be a
statement, or also carry the same request as in example 1, and thus also be an
illocutionary metonym. In example 4, metonymy, again at both the predicational and
illocutionary levels, is combined with ellipsis. In this instance, the request is for the
doctor to agree to chart medication for discretional use. The auxiliary verb ‘charted’ is
elided, but understood.
6 ‘PRN’ stands for ‘pro re nata’ or ‘as needed’ medication, which is charted by doctors for
discretional use by nurses to dispense when they judge that routinely prescribed medication is
inadequate to address the person’s distress. It is typically spoken as an acronym.
Chapter 3 Metonymy
57
From this simple example, it is possible to see that complex pragmatic operations are
occurring for these sentences to be correctly interpreted, yet all of this passes generally
without notice, consistent with an observed characteristic of metonymy (Littlemore
2015). A fifth example, using the same words as examples 3 and 4, can be used
ironically, ‘I need some PRN’ stated by a nurse heading to the pub at the end of their
shift.
Just as metonymy can pass unnoticed in clinical practice, the actual term metonymy
has figured very little in nursing and mental health nursing literature, but the concept
has been addressed by several authors. Menzies Lyth (1959/1988, p. 166) observed
metonymic language use by nurses in the 1950s, in her study of nursing attrition from a
large teaching hospital in London, though she did not specifically use the term:
For example, nurses often talked about patients not by name but by bed
number or by disease or diseased organ: 'the liver in bed 10' or 'the pneumonia
in bed 15.' Nurses themselves deprecated this practice, but it persisted. The
implicit aim of such devices, which operated both structurally and culturally,
may be described as depersonalization or elimination of individual
distinctiveness in both nurse and patient.
Menzies Lyth (1959/1988) identified the linguistic practice as part of a suite of practices
designed to distance nurses from the people for whom they cared. Other practices
included dividing the nursing role into a number of separate tasks that would be
completed by different nurses, and moving nurses between different wards frequently
and with little notice. A member of the Tavistock Institute, using psychoanalytic
methods to inform organisational studies, Menzies Lyth (1959/1988) reported that the
intended aim of these strategies was to thus protect nurses from the anxiety generated
by close contact with human suffering. She pointed out that, not only did the strategies
not protect nurses from exposure to the suffering of others, it also prevented them from
forming the kinds of attachments to patients that would enable them to process the
resultant anxieties. Menzies Lyth (1959/1988) was explicit about the depersonalising
influence of this metonymic naming practice, and its impact on both patients and
nurses.
In contrast, Peplau (1952/1988, p. 124), writing a few years earlier, suggested a
different potential outcome, if nurses were allowed to practise skills in managing the
anxiety of the people they cared for:
Chapter 3 Metonymy
58
There is the possibility that the nurse will develop skill in aiding the patient to
undergo the discomfort and to utilize the energy provided by the anxiety in
identifying and assessing the difficulties in the situation (italics in original).
In this approach, anxiety is explicitly recognised, and seen as a source of energy within
the interaction. Both approaches are echoed in contemporary nursing literature,
including the influential Tidal model (Barker 2001), and approaches that recognise the
emotional impact of nursing (Corley 2002; Freshwater & Stickley 2003).
Nursing provides an ideal context for micro-cultural language use. People encounter
each other regularly in a circumscribed setting with shared goals and clear role
designations. Nursing is often conducted in busy settings, increasing the privileging of
economic communication. It is perhaps not surprising that examples of nursing
language are used to illustrate metonymy in cognitive linguistic texts:
So, when the nurse says 'The gastric ulcer in Room 12 would like some coffee'
(s)he is using the illness (the gastric ulcer) to identify the patient who has it
(Fauconnier 1997, p. 11, italics and parentheses in original).
The vasectomy/herniated disk in 304 needs a sleeping pill [one nurse to
another in a hospital] (Langacker 1993, p. 29, brackets in original).
Health discourse provides an ideal example of ‘pragmatic function mapping’
(Fauconnier 1997, p. 11), bringing together key concepts including the importance of
context, and the high value placed on economy of speech. The examples of nursing
language cited by Menzies Lyth (1959/1988) and the cognitive linguists (Fauconnier
1997; Langacker 1993) are practically identical. It is one of the many examples of
missed opportunities, where authors do not refer to relevant previous work, and
theoretical possibilities are left unnoticed. The motivations underpinning metonymic
language use that were analysed psychoanalytically by Menzies Lyth (1959/1988)
strongly resonate with the kinds of cognitive processes identified decades later by the
linguists, but they do not draw on her conclusions at all. The current study integrates
knowledge from disparate disciplines.
Metonymic language can be used to promote economic information exchanges,
particularly valuable when a clinical environment is busy. Nursing language richly
illustrates the variety of attributes that can be selected by speakers for attention. A
nurse on a busy unit may refer metonymically to a person being cared for by: bed
number, ‘Bed 15’; diagnosis, ‘the cirrhosis in Bed 15’; secondary diagnosis, ‘the PLHIV’
Chapter 3 Metonymy
59
(person living with human immunodeficiency virus); putative diagnosis, ‘just a P.D.’
(personality disorder); symptom, ‘Oh, you’re the two-day fever’; body part, ‘the kidney’;
procedure, ‘I’ve got two ECTs this morning’ ( for person undergoing Electro Convulsive
Therapy); parents’ surname, ‘Baby Smith’; role, ‘Mum at bedside’ (reportedly
documented regardless of gender of attending parent) (Crawford 2014); appointment
time, ‘your 10.15 is getting toey’; or prosthesis, ‘the trachy’ (for person with a
tracheostomy). It is likely that a colleague in the same context will correctly infer who is
being referred to by any of these terms.
Beyond its contribution to fast and economic communication, routine metonymic
language can contribute to the formation and maintenance of shared identity, and
mastering the jargon can be a sign of cultural belonging for the novice practitioner.
However, cognitive linguists are alert to the benefits and perils of metonymic
indeterminacy:
We sacrifice total reliability for the sake of speed and quantity. The price we pay
includes prejudice and the occasional accident (Hudson 2007, p. 517).
While this is a statement about language use in general, there are implications for
nursing. The prevalence of prejudice among nurses has been documented. (Hamilton
& Manias 2006; Hill 2010; Ross & Goldner 2009). This prejudice can be enacted and
reinforced by the language used within the culture of a ward. This can be observed in
mental health practice, where nursing handover has been identified as a process in
which inferences conveyed about the people receiving care through language can
influence staff attitudes and the subsequent care that is delivered (Hamilton & Manias
2006). This is particularly the case with people who have been diagnosed with a
personality disorder, often referred to using vocalised initials, ‘peedee’. The label can
be applied to anyone whose behaviour nurses disapprove of, whether they have the
formal diagnosis or not. A frequent use of the term is ‘Just a P.D.’, with the term ‘just’
highlighting the assessment that the person has no attributes beyond the putative
diagnosis.
Midwives have challenged the language used around women and birth, often using
humour to convey an important message about how paradigms influence the care that
is provided, for instance by stating ‘Women give birth and pizzas are delivered’ (Hunter
2006). While metonymy is not explicitly mentioned, the examples of language given are
metonymic. The implications of this are further examined in work addressing
breastfeeding. Burns and colleagues (2012) demonstrate how, when midwives speak
Chapter 3 Metonymy
60
of facilitating breastfeeding as ‘mining for liquid gold’, they engage in practices that
focus on the midwife getting the breastmilk into the infant, meanwhile reducing the new
mothers to their breasts, or milk delivery units functioning successfully or not. This can
include material practices such as handling women’s breasts without seeking
permission. The language and the conceptualising of the process are metonymic.
An article by Arntfield (2008) casts an interesting light on the influence of the use of
metonymic language by mental health nurses. Arntfield examined the operation of
‘shorthand’ in police communications, and how this leads to the maintenance of
damaging stereotypes. Arntfield (2008) demonstrates how this shorthand is
underpinned by metonymy. Both police and nurses are exposed to traumatic aspects of
human experience, and workplace cultures are developed in part to manage the
resultant anxiety. Within these cultures language choices are pragmatically motivated
by the need for fast and economic communication, which metonymy supports. As a
consequence, both professions must also contend with the implications of the
‘shorthand’ that results, including the reinforcing of stereotyping attitudes toward
particular populations.
Metonymy and mental illness
At the outset of the present study, a search was undertaken for the terms ‘metonymy’
and ‘metonymic’ in key online database search engines, including Embase, Medline,
Psychinfo and Cinahl. No time limits were used in the search, and the databases
typically present published literature dating as far back as the 1980s. One paper which
explicitly addresses metonymy and mental illness, ‘The contribution of metaphor and
metonymy to delusions’ (Rhodes & Jakes 2004), was retrieved. Rhodes and Jakes
(2004), drawing on Lakoff and Johnson’s (1980) theories about metaphor and
metonymy, initially traced a potential metonymic motivation for a delusion expressed by
a participant in their research, specifically that reference to hydrogen stood for super
powers, as a part for a whole. They further stated:
We hypothesize that many incomprehensible and seemingly chaotic one-off
statements might well be a form of unexplained metonymy in that a person only
mentions fragments from a domain (Rhodes & Jakes 2004, p. 7).
Littlemore (2015) reconsidered the data presented in the article by Rhodes and Jakes
(2004), and speculated that other examples that the original authors assessed as
metaphoric could perhaps be better explained as metonymic. Neither study makes
mention of the ‘metonymic speech’ observed by earlier researchers (Cameron 1944;
Chapter 3 Metonymy
61
Goldstein 1944). A repeated search of the databases Cinahl, Psychinfo and Medline
undertaken in December 2014 uncovered no new research into metonymy and mental
illness.
The database searches did not provide any links, historical or contemporary, to the
literature on metonymic speech from the 1930s and 1940s (Cameron 1938; Cameron
1944; Goldstein 1944). I came across the work when I was searching for a definition of
metonymy in a dictionary of linguistics and subsequently sourced the original literature.
LABELLING THEORY AND STIGMA
The powerful link between metonymy and mental illness is forged in the transfer of
meaning from diagnosis as a clinical signifier to diagnosis as a marker of identity. In
simple terms, this is an example of the basic referential part for whole metonymy,
whereby diagnosis stands for identity. Goffman (1963, p. 51) identified how labelling
others with a specific identity on the basis of certain attributes:
… is an offshoot of something basic in society, the stereotyping or "profiling" of
our normative expectations regarding conduct and character; stereotyping is
classically reserved for customers, orientals and motorists, that is, persons who
fall into very broad categories and who may be passing strangers to us.
Cienki (2007, p. 180), a cognitive linguist, notes how metonymy contributes to
stereotyping that can lead to dangerous inferencing:
Social stereotypes are one example of how certain salient members of a
category can be used to represent the entire category (part stands for whole),
with the inherent danger that inferences can be made about the entire category
of people based on the characteristics associated with the one subgroup.
Goffman (1963, p. 5) had already mapped those dangers:
In all of these various instances of stigma however, including those the Greeks
had in mind, the same sociological features are found: an individual who might
have been received easily in ordinary social intercourse possesses a trait that
can obtrude itself upon attention and turn those of us whom he meets away
from him, breaking the claim that his other attributes have on us. He possesses
a stigma, an undesired differentness from what we had anticipated. We and
those who do not depart negatively from the particular expectations at issue I
shall call the normals…By definition, of course, we believe the person with a
stigma is not quite human. On this assumption we exercise varieties of
Chapter 3 Metonymy
62
discrimination, through which we effectively, if often unthinkingly, reduce his life
chances (italics in original).
Goffman (1963) does not use the term metonymy, but he does not need to. The
process he describes, where one attribute’s salience can overshadow all other
attributes, matches the cognitive process of metonymic inference. Goffman (1963)
strategically defines ‘normals’ as ‘We and those who do not depart negatively from the
particular expectations at issue’, ironically rendering the term ‘normal’ one that is
defined contingently and negatively, that is, by what it is not, rather than what it is. He
also highlights how, once a person has been marked by a selected attribute, this give
others the licence to ignore other attributes the stigmatised person may possess.
Goffman (1963, p. 122) further traces elements of the operation of stigma when the line
between ‘normal’ and other is lightly drawn:
The general formula is apparent. The stigmatized individual is asked to act so
as to imply neither that his burden is heavy nor that bearing it has made him
different from us; at the same time he must keep himself at that remove from us
which ensures our painlessly being able to confirm this belief about him. Put
differently, he is advised to reciprocate naturally with an acceptance of himself
and us, an acceptance of him that we have not quite extended to him in the first
place. A phantom acceptance is thus allowed to provide the base for a phantom
normalcy (italics in original).
These processes are inferential. They depend on manipulation of what is allowed to be
foregrounded, and what allowed to recede into the background, that is, the operation of
metonymy. Thus, an attribute typically regarded as salient, such as experience of
mental illness, may temporarily be granted lesser attention, on the proviso that it may
be remobilised at any time, with its full exclusionary power.
Goffman (1963, p. 122) also describes a process how a normally stigmatised person
can be drawn in to questioning the operation of stigma:
He can even be led to join with normals in suggesting to the discontented
among his own that the slights they sense are imagined slights - which of
course is likely at times, because at many social boundaries the markers are
designed to be faint so as to allow everyone to proceed as though fully
accepted, and this means that it will be realistic to be oriented to minimal signs
perhaps not meant.
Chapter 3 Metonymy
63
Again, this process depends on features of metonymy, in particular, the way that two
meanings can be held in play at once. A stigmatising inference can be created, and
simultaneously disavowed, possibly with genuine intent. These are dark and
uncomfortable ideas. I have quoted Goffman at length, partially because of the
influence his work on stigma has had over the fifty years since he published his book,
but also because the link between metonymy and stigma, though implicit, is evident
throughout his work.
The work of Goffman (1963) and Becker (1963) on the labelling of deviance, including
psychiatric labelling, also aligns with cognitive linguistic work on how language is used
strategically by groups to achieve certain interests (Gibbs 2007c). Thus behaviour that
is either tolerated, or even not noted as being different, can be selected for (usually
negative) attention under certain circumstances.7
Zola (1993), in challenging stigmatising language used about people with disability,
outlined a role for the use of grammar. He posited a gradient from the use of a noun to
describe a person with a disability, for example, ‘cripple’, through the use of an
adjective, for example, ‘disabled person’, to the use of a full phrase, ‘person with a
disability’. He argued that the first two uses still place the disability in the foreground of
the reference, but the latter puts the person first. He speculated that the additional time
taken to articulate the full phrase would allow speakers opportunity to reflect on the
person as a whole. While the parallels with cognitive linguistic descriptions of
metonymy are clear, Zola (1993) did not refer to the concurrent work (see, e.g. Lakoff
1987; Lakoff & Johnson 1980).
The recently established National Mental Health Commission has stated the
importance of careful use of language in all their documents, and they use the term,
‘people with lived experience of mental health issues’ (National Mental Health
Commission 2012). This incorporates the earlier work by Zola (1993) in highlighting the
person or people first, and it reflects the recognition that there are people with mental
health issues who have no contact with mental health services, and so cannot be
identified as ‘consumers’.
A further approach to shifting stigma through changing language, with direct relevance
to this study, has been the contentious call for the term ‘schizophrenia’ to be
abandoned (Kingdon et al. 2007). Takahashi and colleagues (2009) report on positive
changes in attitudes held by young people in Japan following the change of name of
7 Goffman notes one counter-example; in a private hospital, patients were referred to by bed
numbers rather than their names as a sign of respect.
Chapter 3 Metonymy
64
the disorder from ‘Mind-Split Disease’ (compare, Fusar-Poli & Politi 2008) to
‘Integration disorder’, however the study is limited as the researchers specifically
compared attitudes to the terms in relation to perceptions of criminality or victimhood.
In a subsequent study, Tranulis and colleagues (2013) compared attitudes toward the
terms ‘schizophrenia’ and ‘salience syndrome’ held by university students, and people
experiencing their first episode of psychosis. They reported that the perceived benefit
of the latter term to the people experiencing psychosis rested on its relative obscurity,
which enabled them to avoid societal stigma in the short term. The authors concluded
that:
The negative stereotypes, separation, power imbalance and discrimination do
not wane simply by changing a label and it is probable that they infuse back into
the new labels as society gets to understand their link with the new labels
(Tranulis et al. 2013, p. 4).
This is consistent with the objectives of the current study, which seek to map the
influence of metonymic language on the experience of people who live with mental
illness, and not presume a simplistic relationship between labelling and stigma.
METONYMY AND NARRATIVE THERAPY
While there is no published literature that addresses a potential role for metonymy in
therapeutic work that was located in the searches, the relevance of metonymy in
Narrative Therapy can be deduced. The role of metonymy in underpinning the practice
of labelling, whereby a selected attribute comes to stand for a person’s identity has
been posited. The externalising conversations developed by White (2007) effectively
reverse this process. In order to do this, the processes rely on a cognitive grapple with
the influence of metonymy, although White did not use the term in his published work.
Cashin and colleagues (2013, p. 33), in describing narrative therapy note that:
Much of the information we utilize is often on the back track, or disattend track,
and no longer part of our conscious awareness. Schemata develop in the form
of narratives that color our perception of the world. Narrative therapy is aimed at
surfacing alternate ways of seeing problems and how the person is positioned
in relation to the problem which in turn influences their ability to move forward in
a less-distressed way.
This demonstrates further alignments between metonymy, as described in cognitive
linguistics, and narrative therapy. The concept that we mobilise information in ways that
Chapter 3 Metonymy
65
are unattended fits with the notion that metonymic language is frequently not noticed.
The description of the influence of narrative schemata maps to the Sapir-Whorf
hypothesis that language influences how we think and experience the world (Lakoff
1987). Narrative therapy works by ‘surfacing alternate ways of seeing problems’
(Cashin et al. 2013, p. 33), which parallels the way that metonymy functions to
foreground salient elements of a concept (Langacker 1993).
There are parallels with other approaches to language and therapy In White’s work that
are not explicitly addressed. The respectful attention to the actual words people use
recalls Peplau (1952/1988) and Meares (2005).8 The validation of ‘experience-near
language’ recalls Goldstein’s (1944) respondent who called green, ‘the color of the
grass in Kentucky’. His recognition of how narrative can be remade resonates with the
words of a key cognitive linguist:
Human beings go beyond merely imagining stories that run counter to the
present story. We can also make connections between different stories, or more
generally, between different and conflicting mental spaces (Turner 2007, p.
378).
While not intended therapeutically, Turner highlights elements of narrative that are
taken up in narrative therapy: the capacity to imagine different stories; and the power to
make connections through narrative, notwithstanding the presence of conflict.
Conclusion
While it is possible to trace recurring themes about language and mental illness across
different discourses, these are not reflected within any one disciplinary discourse.
Cognitive linguists use identical examples to nursing theorists, but the insights each
discipline draws about these examples are not incorporated into the other approach.
Psycholinguists perform multiple observations on the language of people with mental
illness, and speculate about the meanings of their findings, but do not ask their
participants what the experience means to them. Therapists from a variety of
disciplines champion the primacy of people’s own words, but are often blind to
neuropsychological impairments that underpin some of this language use.
Prior to his death, Michael White discussed meeting Russell Meares, and being excited by
some of the parallels between narrative and conversational therapy, and by their shared interest
in the work of James and Vygotsky. The potential collaboration they reportedly planned never
eventuated.
8
Chapter 3 Metonymy
66
There is a gap in our knowledge, as evidenced in the published literature. This study
takes as its focus the topic of metonymy as it is experienced by people with mental
illness. Metonymy has been explicitly addressed within rhetoric, linguistics and
psychiatry. Metonymic influences underpin labelling theory in sociology, and the
resultant stigma people experience. It is also evident in identity formation processes
that highlight certain personal attributes as salient, while letting others recede into the
background. This study explores metonymy in its explicit and implicit forms, through the
words of people who live with mental illness.
Chapter 3 Metonymy
67
Chapter 3 Metonymy
68
CHAPTER 4: METHOD
‘You can’t record auditory hallucinations, put them on a tape’
[Michael, Interview 2, line 375].
Introduction
The previous chapters addressed the concepts underpinning the study, and identified
how the various literatures that were reviewed provided suggestive ways to explore the
topic. The methodological approach to the study was qualitative and interpretive as the
research was aimed at ascertaining what issues, if any, language posed for people with
mental illness, and then exploring any identified issues in detail. There was, however,
no straightforward method described in the research literature that would answer the
research questions.
Data collection was through interviews with people who live with mental illness, and
who had experienced psychosis. Two interviews were conducted with each participant
using the Biographical Narrative Interpretive Method (Wengraf 2001) for data
collection. For analysis of data, two approaches were applied. First, Template Analysis
(King 2004) was employed deductively to code the language participants used within
the interviews. The data thereby coded was synthesised, in order to inductively develop
the overall themes that arose from the interviews.
This chapter describes the reasons for the approaches taken to data collection and
analysis, including the selection and recruitment of participants, the conduct of the
interviews, and the process of analysis. There was a tension implicit in the topic under
study. On the one hand, I aimed to understand the meaning that people make of their
experiences, an approach that must, to some degree, take people’s statements at face
value. On the other hand, I also aimed to analyse their words from a cognitive linguistic
point of view, which posits language choices as frequently occurring unconsciously and
as being influenced by internal and external factors. The research method was
developed in part to manage this tension. The method included a combination of
techniques from oral history, linguistic analysis and template analysis, and incorporated
deductive and inductive analytic strategies.
Chapter 4 Method
69
Participant selection and recruitment
The study was designed to explore the language of people who have had psychotic
illnesses, in particular, how people experienced language when they were in recovery.
Many linguistic phenomena observed in acute psychosis frequently remit once
treatment is initiated (Docherty 2012), and I was interested in the type of language
problems that persisted and influenced people’s everyday lives. Participants’ views on
these experiences were more likely to be accessible when they were not in an acute
psychotic state. The study was designed to elicit information in more depth than simple
confirmation from participants that problems exist. It was also designed to ascertain if
there were discernible patterns of metonymic language use, by and about people who
live with mental illness.
The first inclusion criterion for participation in the study was that the person had
experienced psychosis, as the subject of the thesis was to explore their experiences.
The second criterion was that they had the capacity to provide informed consent. As I
was the sole researcher and conducted all of the interviews, transcribed and analysed
the data, participation was restricted to people who can speak English, which is the
only language I speak.
Participants were recruited through two means. Eight participants were recruited
through clinical contacts within one local health district. The study was advertised to
clinicians through presentations in ‘in-service’ sessions. Clinicians would then approach
individuals who met the inclusion criteria to check if they would be interested in
participating, providing them with the written ‘Information for Participants’ (Appendix A).
The potential participant could then choose to contact me directly, or provide details for
me to contact them. The study was also advertised on two websites accessed by
people with interest in mental health; the New South Wales Consumer Advisory Group,
and the Schizophrenia Fellowship of New South Wales with a brief explanatory text
(Appendix B). Two participants contacted the researcher directly in response to these
advertisements.
Ten participants were interviewed. Nine participants were also interviewed a second
time. All participants were engaged with some form of mental health treatment on a
voluntary basis. None of the participants were receiving involuntary treatment under
mental health legislation. None of the participants were experiencing acute psychosis
at the time of either interview. All participants were taking prescribed anti-psychotic
medication, and several participants were also taking mood stabilising medications.
Chapter 4 Method
70
Data Collection
Interviews were chosen as the method for data collection. Interviews enable the
researcher to ascertain the thoughts and opinions of particular people about a certain
topic (Minichiello et al. 2004; Wengraf 2001). They are ideally suited to exploring
people’s views about their own experience but it is not possible to guarantee that they
provide facts about that experience. For instance, if an interview participant tells of an
encounter with another person, that other person may have a different, and indeed
contested, view of that encounter. However, if the researcher wishes to elicit the
meaning a person attaches to their experience, asking the person is a straightforward
means of achieving this. Interviews enabled the collection of data in the form of
participants’ authentic speech for the purpose of linguistic analysis.
What constitutes acceptable language data has become a contentious issue in the
discipline of cognitive linguistics. Cognitive linguists have been criticised for relying too
heavily on constructed examples of speech rather than empirical data (Brdar-Szabo &
Brdar 2012). Traditionally, most of the writers on metonymy have simply provided
constructed examples themselves. Many of these mirror, or at least approximate,
actual speech with a fair degree of authenticity, but they are constructions. One
approach to address this concern is the study of ‘corpus linguistics’, that is, analysis is
conducted on data from an existing authentic language source (Markert & Nissim
2003). In this approach, examples of context-based language are analysed, often
focused on very specific types of language, including jokes and place names (Feyaerts
2000; Langacker 2009). The data are selected from pre-existing data sources, which
have typically been created for the purpose of analysis of language, but not for
answering a specific research question. In this study, I have conducted interviews and
created a specific data set for empirical analysis.
An elicited narrative is not everyday language. Wengraf (2001) highlights the fact that,
notwithstanding their preparedness to assist the researcher, research participants
invariably bring feelings related to previous experiences of being interviewed to the
encounter. Nonetheless, relating stories is not an atypical use of language in many
domains, so, while telling a requested narrative to an interviewer may be unusual, most
participants will have prior experience of relating narratives to others. FischerRosenthal (2000, p. 116), an historian who participated in the development of the
Biographical Narrative method used in this study, notes:
Chapter 4 Method
71
Empirically, everyone is able to tell their life story, narrate a biography or parts
of it in defined situations… Such situations range from hardly institutionalised
occasional interactions between strangers (in a bar, in a train) to repeated
narration in one's own social milieu (after dinner, during family celebrations), to
highly formalised organisational settings (job interviews, visits to the doctor,
testimony in court, political campaign speeches) (parentheses in original).
While Fischer-Rosenthal (2000) does not specifically address occasions where the
elicited narrative may include discussion of mental health, people who live with mental
illness have almost invariably had experience of delivering their narratives for the
purpose of clinical assessment. A person may be called upon to provide a biographical
narrative on numerous occasions and this demand carries particular pressures for
people who experience mental illness, especially when this includes psychotic
experiences. At a basic level, people cannot always recall episodes when they have
been psychotic, so including these in a narrative is either not possible, or relies on
reports of actions and speech from others. These can be difficult to integrate into ‘selfknowledge’ as they are reported, not recalled. During psychotic episodes, people can
think and act in ways that are uncharacteristic of their behaviour at other times. It can
be difficult to reconcile these experiences for oneself, let alone provide an account to
others that integrates them. This underlines the choice in this study to request
narratives from people in established recovery from acute psychosis.
BIOGRAPHICAL NARRATIVE INTERPRETIVE METHOD
The specific interview method was adapted from the Biographical Narrative Interpretive
Method (Wengraf 2001). The study had a two-fold purpose, to elicit participants’ own
views about language, and to collect data exemplifying their own use of language. The
Biographical Narrative method enabled data to be collected to meet this dual purpose.
The method comprised two interview sessions, the first aimed solely at eliciting
narrative from the interviewee, the second allowing the researcher to directly include
their theoretical concerns with the participant and further explore aspects of the initial
interview that are relevant to the research topic. The approach was designed to create
a body of language data relatively uninfluenced by the researcher in the first instance,
while subsequently allowing for the person to express their own thoughts about the
topic.
Chapter 4 Method
72
INTERVIEWS
Interviews were conducted over nearly three years, from February 2012 until October
2014. Interviews were conducted in a range of settings, including rooms within the
Faculty of Health at the University of Technology, Sydney, private rooms in regional
libraries, two people’s workplaces and the gardens of a drop-in centre for people with
mental illness. Key factors in the choice of locations were that they were convenient for
participants, and not clinical settings.
INITIAL INTERVIEWS
As the name of the method implies, the first interview ideally consisted of a single
question, inviting the participant to tell their narrative. The question was deliberately
couched in non-pathologising language, without use of diagnostic or other medical
terms. At the beginning of the interview, I asked participants a single question, ‘Can
you tell me your story from when you started to notice that things were different or
unusual, up to whatever point of your recovery you wish to go to?’ Additional questions
were asked in a way that prompted the participant to deliver further narrative. More
typical personal interview questions, including requests for participants’ emotional or
evaluative responses to events they narrate, were avoided. I took note of key points
raised by the person that seemed to have particular meaning within their narrative, and
toward the end of the first interview session, requested further narrative in relation to
these points. Wengraf (2001, p. 128) notes:
It is important to use their words-for-feelings - and indeed their words-foranything - rather than use your own, which may mean something different to
them.
Again, the questions in the initial interview were restricted to generating narrative;
seeking clarification about details was reserved for the second interview. While the
intent of these interviews was research-based, not therapeutic, the explicit focus on the
person’s exact words accords similar values to participants’ self-expression as a
number of therapeutic approaches (Meares 2005; Peplau 1952/1988; White 2007).
At the end of the first interview, I had a narrative, in the person’s own words, which
included the events that the participants themselves had selected as personally
meaningful. One of the advantages of this method was that it allowed the interview
participant a significant level of control about what they chose to discuss. I did not ask
questions designed to probe or challenge the participant. This interview approach
created a situation in which the participant could determine what they would address.
Chapter 4 Method
73
Conversely, this also gave a measure of responsibility to the participant to manage
risks to their sense of safety. The method was therefore appropriate to the research
topic, experience related to mental illness, which carried the potential to include
distressing memories for participants.
In most interviews, the number of turns taken by the interviewer and the participant is
controlled almost exclusively by the interviewer, as they are asking the questions. The
first interview in the Biographical Narrative method reverses this power dynamic, as it is
the interview participant who decides how long to speak in response to a question, and
therefore when the interviewer may take their next turn.
SECOND INTERVIEWS
The data from the first interview were provisionally analysed, and a second interview
was typically conducted a month later. The second interview created the opportunity for
me to directly question the participants about issues related to the research topic. My
preliminary analysis of the data from the first interview was discussed with the
participant. At least three types of questions were asked in the second interview. I
presented my interpretation of the data from the first interview, and ascertained the
participant’s views on this interpretation. A second type of question elicited clarification
of specific details, and information relating to the participant’s own views about the
experiences they related. Thirdly, I asked questions related to the research that were
not specifically raised in the first interview. Thus, at the end of the second interview, in
addition to the initial narrative, I had information about the person’s evaluation of their
experience and opinions on the research topic.
The second interviews were conducted between three weeks and six months after the
first interviews for each participant. The variation in these periods relates to external
factors that influenced the availability of the interviewer and the participants, not to
delays in performing provisional analysis of the data from the first interviews.
For each of the second interviews, there were a number of set questions that were
asked of every participant:
Is there anything you would like to change, add or delete from the previous
interview?
Have you had the experience of not being able to follow the speech of other
people?
Chapter 4 Method
74
Have you had the experience of other people not being able to follow your own
speech?
Have you had any experiences where the language used by mental health workers
helped or hindered you?
If I asked you, ‘Would you like another glass?’ what would I mean?
Apart from the first question, which was always asked at the beginning of the second
interview, the questions were not necessarily asked in this specific order, but
interspersed throughout the second interview in order to maintain flow. I also asked the
majority of participants to reflect on two words, ‘insight’ and ‘recovery’, that have
specific and contested meanings within mental health discourses.
The fifth question was introduced as a ‘quick quiz’: ‘If I asked you, “Would you like
another glass?” what would I mean?’ This question was aimed at directly testing what
pragmatic inferences people would draw from a typical metonymic phrase. The
question was designed to ascertain participants’ capacity to comprehend metonymy in
an ‘everyday’ context. It is a prototypical referential metonymy, where the container
stands for the contents. It is an example of metonymy that potentially could be
interpreted literally, that is, the speaker could simply be offering the person a second
glass. However, in most contexts, it is the non-literal meaning of the contents, for
example, water or wine, that is present as the salient feature.
For each participant, there was also a series of questions more directly pertaining to
the first interviews. In particular, examples of their use of language were outlined, and
they were asked if they were aware of this use, and if so, whether this involved
deliberate choice. For instance, one participant consistently used gender neutral terms
when talking about significant people in her life; another used only clinical terms to
describe his experience of psychosis; and a third frequently completed my sentences
before I did, using the words I would have chosen. Each of these phenomena were
explored with participants in the second interviews.
Data Management
In total, fifteen hours of interview data were recorded, which resulted in just under 400
pages of transcript. Interviews were recorded using a digital audio recording device. I
transcribed all of the interviews verbatim. This was undertaken in order to develop
strong familiarity with the material, including how the transcript reflected the tone and
timing of the recordings.
Chapter 4 Method
75
I used a transcription style where passages of text are separated by commas to
indicate slight pauses. This renders the flow of speech more accurately than strict
punctuation. I have not noted small vocalised pauses like ‘um’ and ‘ah’ as in the
passage from recording to text these can inadvertently present the speaker as less
articulate than they are. I have retained all fillers, for example, ‘you know’, ‘like’, and
‘and that’, as they do convey speaker’s style. For the same reason, when a person’s
speech includes grammatical inconsistencies, for example, ‘If I was’, I have not
corrected these. I have used words rather than phonetic approximations, for example,
‘want to’ rather than ‘wanna’, as these prove distracting to readers.
The transcripts were imported into QSR International’s NVIVO 10 (2012), a software
program used for managing qualitative data. Data were initially analysed using the
Template Analysis method (King 2004).
Data Analysis
The first research question in the current study was, ‘How does the use of metonymic
language influence the experience of people who live with mental illness?’ In order to
answer this question, one arm of the analytic process was designed to consider the
data thematically. A number of themes were identified a priori, as I knew I wanted to
explore the issues of identity, stigma, and people’s experience of language, particularly
their experiences of language encountered from mental health workers. These themes
have been identified in the literature (see, e.g.Barnes & Berke 1971; Brown &
Kandirikirira 2007; Goffman 1968; Lysaker, Roe & Yanos 2006; Ross & Goldner 2009;
Shefer et al. 2014) and I was interested to know if they were salient in participants’
lived experience. This entailed deductive analysis of the data to identify if these themes
were present. Hammersley (2008, p. 69) notes, ‘it is impossible to begin research
without prior assumptions of some kind’. However, the study was intended to be both
interpretive and explorative, and an integral part of this was to analyse the data
inductively as well. This entailed observing for recurrent themes arising in the
narratives, and also noting ideas or experiences that weren’t necessarily recurrent, but
cast the processes in a different light.
The second arm of the analytic process was designed to investigate the language in
which participants conveyed their narratives, in order to answer the second research
question: ‘How do metonymic patterns influence the speech production and speech
comprehension of people who live with mental illness?’ This entailed examining the
language for cognitive linguistic, clinical and narrative features. The analytic method
Chapter 4 Method
76
thus combined several approaches that were to run concurrently, prior to being
synthesised. Template analysis was selected as the method that enabled me to
maintain organisation of the disparate streams of analysis, but also make them
available for synthesis.
TEMPLATE ANALYSIS
Data were deductively analysed using a series of templates. This method has been
described by King (2004, p. 256) who states that:
The term ‘template analysis’ does not describe a single clearly delineated
method; it refers rather to a varied but related group of techniques for
thematically organizing and analysing textual data.
Typically, thematic analysis involves a process of coding ‘chunks’ of data that fit certain
themes that have been identified in advance as relevant to the research question(s)
(Wengraf 2001). In template analysis, these themes are literally organised in a
template, so the researcher is referring to the pre-existing codes as they analyse (King
2004). The data are then analysed, coding for these themes, but also being alert to
themes arising from the data. The template was revised to reflect themes that emerge
from the data, and also to remove initial codes that were not reflected in the data. A
final template was then set and all the data re-analysed. Coding the data using the final
template does not represent the final step in analysing the data – the templates exist
primarily to help organise the data and support analysis.
King (2004, p. 258) flagged the potential for parallel coding, ‘whereby the same
segment is classified within two (or more) different codes at the same level’. My use of
template analysis introduced an adaptation from the beginning. Template analysis
typically codes data for themes only, arranging these in hierarchies in order to
synthesise the analysis. In addition to thematic analysis, I conducted linguistic analysis
of the data, and while this intersected with the thematic analysis, it was not
incorporated into a hierarchical structure for synthesis, but remained distinct.
TEMPLATES
I analysed the data using four templates; one thematic, one identifying language
practices described within cognitive linguistics, one using terms from the Thought,
Language and Communication Scale (Andreasen 1979a), and one using Wengraf’s
‘textsort’ categories (2001). Parallel coding occurred within templates, for instance, a
segment was coded as evincing the themes of ‘Experience of language’ and
‘Experience of treatment’. Parallel coding also occurred across templates, for instance
Chapter 4 Method
77
a segment was coded thematically as ‘identity’, and also as an instance of ‘metonymy’
from the cognitive linguistic template.
An initial a priori set of codes were established within each template, capturing the key
elements of the research questions. I knew that I would be looking for examples of
metonymy, and other aspects of language identified in cognitive linguistics. I also knew
that I would be considering certain thematic aspects, including identity, the experience
of symptoms, and experiences related to language, especially in encounters with
health workers.
These initial codes were trialled using some of the data, then further refined to reflect
information arising from the data. A number of codes were added, for example, I had
not anticipated the extent of use of idiomatic language, and so established a code to
track this. Codes were also deleted, after preliminary analysis revealed they were not
substantially present in the data. A final set of templates was constructed, and all data,
including that which had been provisionally analysed, were coded to this set of
templates.
TEMPLATE 1: THEMES
The data were analysed thematically. The template of codes for themes underwent the
most change of the four templates following preliminary analysis of the data. This is
consistent with exploratory qualitative data analysis, where themes frequently emerge
from analysis of data. I had known I wanted to investigate several themes from the
outset of the study. These included the theme of identity and how people formed their
sense of self. I deliberately did not link this to explicit references to the influence of
mental health diagnosis on identity formation, as I wished to elicit the broadest ideas
people had about identity, and understand what elements they themselves selected in
this process. Indeed, I did not even wish to presume that identity formation was a
salient idea for participants. I knew that I would be looking for examples of stigma. I
also knew I would be investigating people’s own words about their experiences of
language, and their experiences of treatment, though here again, I did not want to
explicitly link the two processes, at least in the first interview.
A number of earlier thematic topics included resilience, the experience of psychosis
diminishing, and coping mechanisms. These had been suggested both by my clinical
experience, and from the literature (Brown & Kandirikirira 2007; Deveson 2003; Slade
2009). However, in response to analysis of the data, the concept of recovery better
Chapter 4 Method
78
captured these. Another topic that was added was ‘insight’ as this came up repeatedly
in the narratives. The final set of themes that were coded is set out in Table 4.1.
Table 4.1 Themes
Term
Definition
Experience of
symptoms
Experience of
treatment
Experience with
language
Experience of symptoms of mental health issues, including positive
and negative experiences
Experience of treatment by health professionals, including mental
health professionals
Notable experiences of language. Includes differences in speech
production and comprehension, as well as experiences of language
used about them
Participants’ expressed thoughts about identity
Both participants’ expressed thoughts about the term, as well as
reported experiences of developing insight
Both participants’ expressed thoughts about the term, and reported
experiences of recovery
Experiences of overt or perceived discrimination related to mental
illness or other attributes. Includes self-stigma
Identity
Insight
Recovery
Stigma
TEMPLATE 2: COGNITIVE LINGUISTIC TERMS
The research questions that guide the study focus on metonymy as it is defined in
cognitive linguistics. Metonymy is not the only process attended to within this field of
study and it frequently occurs in combination with other figures of speech. A number of
salient features were identified. These are listed and defined in Table 4.2. These were
mostly set before the data were analysed, as they were key concepts guiding the
study. Idiom, irony and understatement were added after the preliminary analysis of the
data, as they recurred frequently. As metaphor has been addressed widely in
therapeutic and cognitive linguistic literatures, I decided only to include it when it was
particularly notable.
Chapter 4 Method
79
Table 4.2 Cognitive linguistic terms
Term
Blend
Ellipsis
Idiom
Irony
Metaphor
Metonymy –
referential
Metonymy –
predicational
Metonymy –
illocutionary
Simile
Understatement
Definition
Blending is in principle a simple operation, but in practice gives rise
to myriad possibilities. It operates on two Input mental spaces to
yield a third space, the blend. The blend inherits partial structure
from the input spaces and has emergent structure of its own
(Fauconnier 1997, p. 149)
The omission from speech or writing of a word or words that are
superfluous or able to be understood from contextual clues (Oxford
Dictionaries 2015)
An idiom is a linguistic form consisting of two or more morphemes
in a construction whose overall meaning is not derivable by
compositional means from the meanings of the component
morphemes in that construction (Talmy 2007, p. 276)
A form of speech in which one meaning is stated, and a different,
usually antithetical, meaning is intended (Preminger, Warnke &
Hardison 1986, p. 109)
A form of conceptual projection involving mappings or
correspondences holding between distinct conceptual domains
(Evans 2007, p. 136)
Metonymy in which a source term is used to indirectly refer to a
target object or an entity, and the terms are linked by conceptual
contiguity.
Metonymy in which the substitution of source for target addresses
the properties of an object, rather than the object itself
Metonymy in which a speech act stands for an indirect request
A form of metaphor where the link between concepts is specified
as ‘like a…’ or ‘as a…’
The presentation of something as being smaller or less good or
important than it really is (Oxford Dictionaries 2015)
Blends, Idioms and Understatement all frequently depend on metonymy to function,
either as motivating factor, or in combination with metaphor. This confirmed Gibbs’
(2007a, p. 697) assertion that, ‘In fact, it is nearly impossible to speak of many human
events without employing idiomatic phrases that communicate non-literal meaning’.
The coding of cognitive linguistic terms posed specific challenges. The disagreement
among linguists about the borders between different terms was noted in the chapter on
Metonymy, and recently restated by Littlemore (2015) in her monograph about the
topic. Moreover, as Barthes (1994, p. 86) notes:
Yet we still lack (but perhaps such a thing is impossible to produce) a purely
operational classification of the principal figures: dictionaries of rhetoric, indeed,
Chapter 4 Method
80
permit us to discover what a chleuasmus is, or an epanalepsis, or a paralepsis,
to proceed from the often quite hermetic name to the example; but no book
allows us to make the converse trajectory, to proceed from the sentence (found
in a text) to the name of a figure; if I read: so much trembling over so much
shadow, what book will tell me this is a hypallage, if I do not already know this?
We lack an inductive instrument.
In the absence of such an inductive instrument, the transcripts were considered
multiple times, and examples of key figures subjected to repeated analysis prior to
inclusion in the results.
TEMPLATE 3: ANDREASEN’S THOUGHT, LANGUAGE AND
COMMUNICATION SCALE
In tracking other notable aspects of language use, the Thought, Communication and
Language Scale first developed by Andreasen (1979a) was used as another template
to enable reliable differentiation of the data. While the scale has explicit diagnostic and
prognostic intent that is not part of this particular study, it is used here as it contains
operationalised definitions of language patterns observed in the speech of people who
live with mental illness. These are set out in Table 4.3 below. As many of the terms had
arisen in the first pass through the data, it was simple to adopt the scale, and there is
an advantage in discussing the findings in using a widely cited scale. The term ‘scale’
is slightly misleading. Andreasen, after providing operationalised definitions for the
aspects of language, also developed a numeric scale, with items variously ranged from
0 to 4 or 5, in terms of their frequency of use. In the event, in all published articles by
the author and colleagues (Andreasen 1979a, 1979b; Andreasen & Grove 1986;
Docherty 2012), any item that rated a 1 or above, that is, occurred at least once in the
data, was counted for the purposes of analysis, thus rendering the numeric scale
redundant.
Chapter 4 Method
81
Table 4.3 Andreasen's TLC Scale
Term9
Blocking
Circumstantiality
Clanging
Derailment
Distractible speech
Echolalia
Illogicality
Incoherence
Loss of goal
Neologisms
Perseveration
Poverty of content
of speech
Poverty of speech
Pressure of
speech
Self-reference
Stilted speech
Tangentiality
Word
approximations
(Paraphasia,
metonyms)
9
Definition
Interruption of a train of speech before a thought or idea has been
completed.
A pattern of speech that is very indirect and delayed in reaching
its goal idea. In the process of explaining something, the speaker
brings in many tedious details.
A pattern of speech in which sounds rather than meaningful
relationships appear to govern word choice, so that the
intelligibility of the speech is impaired.
A pattern of spontaneous speech in which the ideas slip off the
track onto another one that is clearly but obliquely related, or onto
one that is completely unrelated.
During the course of a discussion or interview, the patient
repeatedly stops talking in the middle of a sentence or idea and
changes the subject in response to a nearby stimulus.
A pattern of speech in which the patient echoes words or phrases
of the interviewer.
A pattern of speech in which conclusions are reached that do not
follow logically.
A pattern of speech that is essentially incomprehensible at times.
Failure to follow a chain of thought through to its natural
conclusion.
New word formations. A neologism is defined here as a
completely new word or phrase whose derivation cannot be
understood.
Persistent repetition of words, ideas or subjects so that, once a
patient begins a particular subject or uses a particular word, he
continually returns to it in the process of speaking.
Although replies are long enough so that speech is adequate in
amount, it conveys little information.
Restriction in the amount of spontaneous speech, so that replies
to questions tend to be brief, concrete and unelaborated.
An increase in the amount of spontaneous speech as compared
with what is considered ordinary or socially customary.
A disorder in which the patient repeatedly refers the subject under
discussion back to himself when someone else is talking.
Speech that has an excessively stilted or formal quality.
Replying to a question in an oblique, tangential, or even irrelevant
manner.
Old words that are used in a new and unconventional way, or
new words that are developed by conventional rules of word
formation.
All of the terms and definitions in Template 3 are from Andreasen (1979a, pp. 1318-21)
Chapter 4 Method
82
TEMPLATE 4: NARRATIVE MODES: WENGRAF’S ‘TEXTSORT’
CATEGORIES
A final template was used to code the transcripts into the ‘textsort’ categories outlined
in the Biographical Narrative method, as described by Wengraf (2001). These
categories are set out in Table 4.4. They are intended as a means of distinguishing
different narrative practices that participants used within an overall narrative. However,
there are not strict boundaries between the categories; they are suggestive rather than
fixed. The advantage of the textsort categories is that they are more flexible than
standard narrative divisions, in which the parts of narrative structure are determined by
their temporal sequence in the overall narrative, for example, abstract/ orientation/
complicating action/ evaluation/ resolution/ coda (Labov & Waletzky 1967/1997).
Table 4.4 Wengraf's ‘textsort’ categories
Term10
Definition
Description The assertion that certain entities have certain properties, but in a
timeless and non-historical way
Argument
The development of argument and theorizing and position-taking, usually
from a present-time perspective
Report
A sequence of events, experiences and actions is recounted, but in a
relatively experience-thin fashion
Narrative
The telling of a story by which event Y followed event X, and event Z
followed event Y, either for causal reasons or just ‘because they did’. The
story is not told in a very ‘thin’ way…but rather in ‘rich detail’
Evaluation The ‘moral of the story’ stated explicitly as such.
Even with this relative flexibility, the interviews were not able to be coded strictly to the
model outlined by Wengraf (2001), as most participants moved between narrating
styles more frequently than was consistent with the model. Nonetheless, these shifts
between textsort categories did occasionally correlate with shifts in other linguistic
features, for example, higher density of figurative language was noted when
participants were using a ‘report’ mode in the textsort categories, than when they were
using an expanded ‘narrative’ mode.
SYNTHESIS OF TEMPLATE ANALYSES
Subsequent analysis consisted of examining intersections between coded items, to
identify patterns and trends. Some of these were flagged in the research questions, for
example, metonymic use of diagnosis (from the cognitive linguistic template), and
10
Terms and definitions are adapted from Wengraf (2001, pp. 243-244)
Chapter 4 Method
83
identity (from the themes template). Patterns that characterised the speech of
individual participants were also observed. Synthesis of the data also persisted through
the interpretation of results, as themes and linguistic features emerged in the process
of setting out results in a communicable fashion.
Ethical considerations
The study was initially reviewed by the Human Research Ethics Committee (HREC) of
Sydney Local Health District and approval granted (Reference X11-0031 &
HREC/11/RPAH/41). This approval was subsequently ratified by the University of
Technology HREC (Reference 2011-109R).
The research was guided by The National Statement on Ethical Conduct in Human
Research (2007/2015) which states that, ‘People with a cognitive impairment, an
intellectual disability, or a mental illness are entitled to participate in research, and to do
so for altruistic reasons’ (National Health and Medical Research Council, Australian
Research Council & Australian Vice-Chancellor's Committee 2007/2015, p. 58). A key
aspect of the current research was to give voice to the experience of people who live
with mental illness in relation to an issue where they have generally been treated as
research subjects, rather than participants, and their language performance assessed
‘objectively’ by researchers. At the same time, and in consideration of, ‘participants’
distinctive vulnerability’ (National Health and Medical Research Council, Australian
Research Council & Australian Vice-Chancellor's Committee 2007/2015, p. 59), care
was taken to minimise the risks to participants.
INFORMED AND VOLUNTARY CONSENT
Capacity to give informed consent was established through direct discussion about the
study with candidates, checking that they understood the benefits and risks prior to
signing a written consent form (Appendix C).
Participation in the study was voluntary. I did not directly recruit the participants, in
order to reduce the potential for coercion, or perceived coercion for people to
participate. I had worked in the local health district from which participants were
recruited, and had previously met four of the participants; two of these people had no
clinical contact with me, and the other two had participated in group activities that I
coordinated, but had never received individual treatment from me. Participants were
not offered compensation for their interviews. In several cases I offered to pay for
people’s transport to the interview locations, but none of the participants accepted this
offer.
Chapter 4 Method
84
RISK AND BENEFIT
The discussions I held with all participants prior to requesting their informed consent to
participate in the study included consideration of the risks and benefits that would
potentially arise from their participation. The primary risk was that narrating their stories
could include recalling traumatic events, and that this would lead to distress. All
participants reported awareness of this risk. The measure of control held by the
interviewee in the Biographical Narrative method partially mitigated this risk, and
beyond the request for their story, I did not ask questions that probed potentially
distressing topics. Participants were also aware that there would be no direct benefit to
them through participation in the study. Several participants explicitly voiced their
understanding that their participation could potentially lead to the generation of
knowledge about the experience of language of people who live with mental illness.
I was aware of the privilege of being granted participants’ narratives of their experience
for the purpose of my research. While maintaining participants’ confidentiality, I have
also attempted to respond to this privilege by treating their stories with respect. The
process of selecting excerpts from people’s narratives is the researcher’s responsibility,
and there is the possibility that participants may consider my selections as not
presenting the most salient examples of their experience, or my interpretations as not
truly representative of their thought. Several participants expressed recognition that this
may occur, and stated altruistic intent as motivating them to participate in research that
may contribute to improvements that would benefit others.
PARTICIPANT DISTRESS
The request to narrate previous experience can bring up unexpected emotional
responses in anyone. For people with experience of mental health issues, there is often
a history of trauma, either as a result of symptoms themselves, through service system
related events, or through negative experiences, including being stigmatised by other
people, including family members, employers or strangers. As a mental health nurse, I
had over fifteen years’ experience of responding to people in distress. Throughout the
interviews, I remained attuned to people’s responses.
With three participants I paused the interviews to suggest to people that they did not
need to continue with the interview, as they were expressing distress as they recalled
past experiences. One participant reported it was traumatic to recall an incident that
occurred during his first hospital admission. Another participant reported that she
generally avoided thinking back on certain experiences, as she felt such thinking
hampered her moving forward. A third participant evinced significant distress when
Chapter 4 Method
85
recounting certain past events, but respectfully declined the interviewer’s offer to stop
the interview, stating that the emotions were congruent to what she was discussing,
and she was able to manage her distress.
In each instance, participants opted to continue, though two participants shifted from
the topics they had been discussing. I did not attempt to counsel any of the
participants, but offered to arrange counselling if they felt they may benefit from it. No
participant took up this option. At the beginning of the second interview, I checked with
each participant if there was anything they wished to add, alter or delete from the
previous interview, and none did.
One participant expressed reluctance to participate in the second interview, reporting
that the initial interview experience had brought issues to his mind that were
distressing. In discussion, it came to light that he had planned to participate in the
second interview despite this reluctance, as he stated he did not wish the information
he had provided in the first interview to be lost to the study. When he was reassured
that information from the first interview could be included in the study without his
participation in a second interview, he expressed relief and opted not to return for the
second interview.
CONFIDENTIALITY
Care has been taken in presenting information provided by participants in the chapters
on Results and Discussion. All data were coded with a participant number, converted to
a pseudonym for the purposes of reporting. Pseudonyms have been used for the
participants, to protect the identity but not interfere with the readability of the results.
Information that may be used to identify participants or people to whom they refer has
been de-identified. Participants may be able to identify elements of their own narrative,
or direct quotations they have made, but other readers will not be able to link
information from the study to particular individuals. Care has been taken when a
narrative includes sensitive information about a third person; this has typically been
excluded from the reported data, in order to avoid the potential for inadvertent
identification of a person. All references to proper names that may be used to identify
people or organisations were removed from the transcripts. Written transcripts of the
interviews were kept in a locked cabinet. Electronic data, including audio recordings
were stored in password protected files on a personal laptop, and on a password
protected external hard drive.
Chapter 4 Method
86
Reflexivity
I worked as a mental health nurse for over fifteen years, with the majority of these
spent in direct clinical contact with people with a range of mental health problems. Over
the course of the study I changed my employment position more than once, moving
from direct clinical contact with people who live with mental illness, to consulting with
them for the purpose of policy development.
I had been thinking about some of the ideas that underpin the present study for several
years prior to beginning. I had studied mental health nursing and counselling,
undertaken workshops in Narrative Therapy with Michael White, and attempted to
integrate these ideas into my clinical practice. I therefore came to the study with certain
ideas already, as well as a set of values about working with people who live with
mental illness within the public mental health system.
Within my experience as a mental health nurse, I had conducted many interviews with
people with a range of presenting problems. These interviews largely fell into two
types; assessments and therapeutic encounters. In the first type, assessments, I was
asking questions in order to formulate clinical impressions, and develop plans with the
person about what kind of intervention might help them. In the second type of interview,
therapeutic encounters, I would ask questions in order to evaluate their current mental
state and response to previous interventions, and formulate ongoing plans with people
about how they may achieve and maintain improvements in their mental health, again,
in many contexts and with people experiencing a range of difficulties. These questions
would frequently incorporate my professional responses to the information I was
hearing.
Conducting interviews as a researcher with people who experienced mental illness
required that I relinquish both of these long-practised approaches. It was important that
I not introduce the ‘fact-finding’ intent of the assessment interview, as this would distort
my capacity to gather participants’ narratives in their own words. It could also have
generated unease in participants, reminding of them of clinical encounters, and
undermining their sense of contributing to research. Nor was I interviewing in order to
practice therapy with participants. This would have represented a blurring of my
professional boundaries.
This dual need, to practise new interviewing skills while keeping others in check, was
challenging at times. I had to learn how to convey empathy to participants as a
researcher, which is a key part of conducting a successful qualitative interview, while
Chapter 4 Method
87
not couching this in terms of my professional knowledge. This was particularly
challenging when participants directly requested my clinical opinion on matters they
were discussing; when this occurred, I would respond with general comments, and
refer them back to their treating teams for detailed discussion. One way in which I
managed these issues was to discuss the challenges they presented with my principal
supervisor.
I also kept a journal throughout the interviews, where I noted my observations of
elements that contributed to building and maintaining rapport with participants, which
questions stalled, and which created flow and cohesion. I referred to these notes when
I was preparing for second interviews with individual participants, and to guide my
overall research interviewing style.
The study presented me with other challenges beyond adopting a researcher stance
rather than a clinical one. A risk in qualitative research is that the researcher only finds
what they set out to find (Hammersley 2008). This risk is compounded when the
researcher has a very specific familiarity with the topic under consideration.
Some of the steps that I took to address this issue involved the establishment of the
study, while some of them are related to the analysis. In the first instance, I made a
decision to conduct interviews with people with whom I had not had direct clinical
contact, the majority of whom I had not met prior to conducting the study. This reduced
the possibility that people would have already heard my particular thoughts about
language, or been influenced by my ideas directly through clinical encounters.
Collecting data in the form of extended narratives also created a space wherein
information I could not anticipate would be provided by participants. Following the
Biographical Narrative method, I had minimal influence on what people chose to
discuss in the initial interviews. Thus, a significant proportion of the data on which I
based my analysis was outside my control, and therefore less likely to be subject to
bias through the introduction of topics or questions.
The second set of steps related to the analytic method. There are the themes I was
particularly interested in when I began the study, including the role played by
metonymy in the formation of identity, and the production of stigma. I was transparent
about looking for these, and they were part of the deductive template analysis aspect of
the study. During analysis of the data, themes I had not anticipated recurred, and I
recorded these and reviewed all the data several times to check on them. This forms
Chapter 4 Method
88
the inductive aspect of the study, and the two approaches are synthesised in the final
report.
Beyond the themes, which were set, either deductively or inductively, by the
researcher, I also conducted linguistic analysis of the data, using terms described in
the cognitive linguistic literature, and also terms from earlier research into language
and psychosis. These analyses strengthened the study, creating data that can be
reviewed independently of the thematic analysis, but can also be used to support the
perceptions of experience reported by participants.
Conclusion
In conducting the study, I was seeking to accomplish a number of things. I wanted to
explore the views of people who live with mental illness about their experience of
language. I also wanted to analyse their language, to ascertain if there were distinctive
patterns that contributed to the way they experienced language. There was no
documented method that would allow me to explore the research questions set out in
the Introduction. In consultation with my principal supervisor, I developed a hybrid
method, which combined a data collection process that created two types of response
to the research questions with an analytic process that allowed for a number of
different approaches to the data to be maintained and synthesised. The results of the
study are detailed in the next chapter.
Chapter 4 Method
89
Chapter 4 Method
90
CHAPTER 5: RESULTS
This chapter sets out the results of analysis of the information provided by participants
in the interviews. The first section of the chapter summarises information about the
participants, including the mental health diagnoses they have and the types of mental
health treatment they have received. The second section describes the interviews
themselves.
The third section of the chapter presents the key themes identified in the data analysis
of the interviews. These are described thematically, in order to reflect the experiences
participants report across varied domains. The specific themes that are addressed are:
identity, stigma, the experience of symptoms, the experience of treatment, and insight
and recovery. Information from the analysis of the other templates, including cognitive
linguistic terms and narrative styles, is incorporated into the thematic results. This is
consistent with the way in which these elements function, underpinning the thematic
expressions.
The final section of the chapter presents analysis of the data about participants’
experience of language. This includes examples of typical and atypical language use
demonstrated by participants, as well as their expressed views on their experience of
language. The sections help to organise the data, but they are not rigid. Where there is
an overlap between topics, this is identified.
The practice of pragmatic inferencing was evident throughout the data, with
participants describing both situations where they were making inferences about other
people’s meanings (speech comprehension), or where they perceived inferences being
made about them (speech production). These events occurred not just when attention
was specifically drawn to language use, but across other domains, including the
formation and reassessment of identity, the experience of being subject to stigma, and
engagements with mental health workers. Metonymic language contributed to this
pragmatic inferencing. Where this occurred, it is described.
Chapter 5 Results
91
Participants
Ten participants were interviewed in the study. Table 5.1 summarises demographic
data about the participants, who have been given pseudonyms.
Table 5.1 Participant characteristics
Identifier
Age
Completed Diagnosis
years of
education
Michael
32
15
Schizophrenia
Years since
initial
contact with
MH services
13
Ethan
58
16
Schizophrenia
40
Tanya
27
10
Schizophrenia
10
Oliver
27
13
2
Nathan
22
16
Schizoaffective
disorder
Depression with
psychotic features
Yves
47
14
Schizophrenia
30
Matthew
49
13
2
Schizoaffective
32
disorder
Ian
43
13
Schizophrenia/
20
Schizoaffective
disorder /Bipolar
disorder*
Ellen
42
16
Bipolar disorder/
13
Post-traumatic
stress
Shirley
40
18
Bipolar disorder/
16
schizoaffective
disorder
* Ian’s diagnosis was under review at the time of the interviews.
Current
vocational
status
Working
part-time/
studying
Working
full-time
Not
working
Unknown
Working
part-time/
studying
Not
working
Not
working
Working
part-time
Not
working
Working
part-time/
studying
Three of the participants were female, seven male. The participants ranged in age from
22 to 55. All participants spoke English. Two participants spoke another language at
home to older family members, though both were born in Australia, and had spoken
English and their parent’s first language from childhood. Two other participants grew up
in environments that were culturally diverse from the dominant Anglo-Celtic Australian
heritage, including one person originally from the Torres Strait Islands.
Nine of the ten participants had completed high school. One participant left school at
15. Six participants had engaged in university education, with three of these completing
Chapter 5 Results
92
undergraduate degrees and one completing more than one degree. Another participant
had completed studies within Technical and Further Education (TAFE).
One participant was working full-time. Four participants were working part-time, with
three of these combining part-time work with part-time study. Four participants were
neither working nor studying currently. One participant’s occupation was not stated.
FORMAL MENTAL HEALTH DIAGNOSES
Past experience of psychosis was an inclusion criterion for participation in the study,
and all participants had formal diagnoses that incorporated psychosis. Four participants
had a formal diagnosis of schizophrenia, two were diagnosed with schizoaffective
disorder, two had diagnoses of bipolar disorder and one had a diagnosis of severe
depression with psychotic features. One person’s diagnosis was under review at the
time of the interviews. Notwithstanding these formal diagnoses, several participants
reported discussions with their treating teams in relation to the applicability of
differential diagnoses, and these are reported in further detail below.
TREATMENT
All participants were currently engaged with voluntary mental health treatment at the
time of the interviews. All participants were currently taking psychotropic medications;
some were taking antipsychotics alone, while others were taking antipsychotics in
combination with anti-depressants or mood-stabilising agents. Seven participants were
also engaged in some form of counselling with one or more mental health workers.
These included their treating psychiatrists, but also mental health nurses,
psychologists, social workers, and support workers.
Nine of the participants reported they had been subject to involuntary mental health
treatment under mental health legislation at some time in the past. The number of
years since participants first had contact with mental health services ranged from two
to 40, with an average length of 17 years. These figures do not strictly correlate with
years in treatment, as several participants have had intervening periods since first
contact without being on any treatment.
All participants had received treatment in the public mental health system. Four
participants had been treated exclusively in the public mental health system, while six
had been treated in both the public and private systems. Three participants were
currently treated exclusively in the private sector, while seven maintained engagement
with public mental health services.
Chapter 5 Results
93
Interviews
There was a marked range in the duration of the interviews, and the number of turns
taken by the participants. These are summarised in Table 5.2.
Table 5.2 Interview data
Identifier Interview 1
Duration
Turns
Interview 2
Turns Interview 1
Longest
taken
Duration
taken
Initial
free
answer
passage
(lines)
(lines)
(minutes)
(minutes)
Michael
31
15
43
58
104
104
Ethan
26
8
39
24
91
91
Tanya
24
34
30
46
4
17
Oliver
12
19
N/A
N/A
N/A*
24
Nathan
35
18
26
35
175
175
Yves
53
40
43
79
57
57
Matthew
34
177
27
122
2
5
Ian
54
141
61
120
15
22
Ellen
92
26
73
45
55
463
Shirley
90
15
101
51
328
328
* Due to a recording error, some minutes from Oliver’s first interview were lost.
There is no distinct pattern in the differential durations between first and second
interviews for each participant, with five of the latter interviews lasting longer, and five
being shorter. The average first interview lasted 44 minutes, and the average second
interview lasted 43 minutes and 45 seconds.
Turns were contained as pairs, that is, 15 turns indicates that the interviewer and
participant each spoke 15 times. Duration of spontaneous speech was measured by
the number of lines in the transcript as this most accurately reflects the amount of
content of speech. The alternate measure of duration in time measured in minutes is
prone to variation in rate of speech, and so does not capture the level of content in
each passage of speech.
One marked variation across participants was the number of turns taken by the
interviewer and the participant. One interview [Ethan, Interview 1], consisted of just 8
turns each for the interviewer and participant, but lasted 26 minutes and generated 227
lines of transcript, of which only 12 lines were spoken by the interviewer. Similarly,
Chapter 5 Results
94
another interview [Shirley, Interview 1] consisted of 15 turns over 90 minutes, with the
interviewer generating 28 lines of transcript and the participant 856. At the other end of
the spectrum, another interview, [Matthew, Interview 1] lasted 34 minutes, but
consisted of 177 turns, and generated 474 lines of transcript, with many of these lines
consisting of just a few words.
There was distinct variation in response to the requested narrative in the initial
interviews. Two speakers, Nathan and Shirley, were able to speak at length (175 and
328 lines respectively) in response to the initial question. In contrast, two other
speakers, Tanya and Matthew, responded with two- and four-line answers respectively.
The length of initial responses did not correlate with the length of responses throughout
each interview. For instance, Ellen’s initial response comprised 55 lines, but later in the
interview she spoke interrupted for approximately 40 minutes (463 lines of transcript),
which was the longest spontaneous passage of speech in any of the interviews.
Matthew’s responses remained brief throughout, with no individual passage in the first
interview exceeding four lines.
The second interviews were semi-structured, with a schedule of questions. For eight of
the ten participants, this resulted in interviews that consisted of significantly more turns
than the first interviews. Two participants’ second interviews comprised fewer turns
than their first (Mathew and Ian), however in both cases, the first interviews comprised
many more turns than taken by the other participants.
Themes
The data were analysed thematically, with a focus on the role played by language
across different aspects of participants’ experience. These themes were: identity;
stigma; experience of symptoms; experience of treatment; insight; and recovery.
IDENTITY
All participants reported that they had thought about their identity, and the relation of
their experience of mental illness to their identity. None of the participants reported this
thinking to be simple or straightforward. Participants reported that the experiences of
psychosis altered their sense of reality. And they also reported that factors that
accompanied the psychosis further contributed to this shift. For all of the participants,
the onset of illness disrupted their anticipated life path. This affected every aspect of
their lives, including relationships with friends, family and lovers, their educational and
vocational ambitions, their financial capacity and their social status. Each of these
aspects can stand as salient attributes in the formation of identity. So, in addition to
Chapter 5 Results
95
having to incorporate the experience of mental illness into their sense of self,
participants also had to make adjustments across many of the other factors that may
have previously contributed to their identity. For some participants, relinquishing these
other factors was as difficult as incorporating mental illness into their identity.
Participants spoke about how their sense of identity was influenced by the highlighting
of certain attributes. The experience of mental illness figured as a key attribute that
frequently altered people’s sense of identity. In contrast to many other factors that
typically shape identity, mental illness was reported as not being voluntarily selected.
Shirley described how being given a diagnosis of mental illness is equated to having
your future determined:
‘Identity always comes up, because it is such an affront to your self, to your
sense of self, to be given one of these words and told that’s your, you know,
and especially with schizophrenia, that’s your destiny, and I’ve had doctors, and
this, this wasn’t helpful to me, but, and I found it quite insulting, but a doctor did
say to my mum at the time, “Oh, she’s got bipolar, you’re lucky she doesn’t
have schizophrenia”’ [Shirley, Interview 2, lines 539-544].
Shirley made explicit the fact that messages that affect the impact of a diagnosis of
mental illness on a person’s identity are directly sent by mental health workers. She
reported her own experience of this type of message as negative. For Shirley, it was
through the specific naming power of language, ‘to be given one of these words’ that
this process occurred.
Ian also spoke of the dramatic shift in his identity following diagnosis with a mental
illness:
‘Like I was a straight-A student up until then, 14 years of age, and after that I
became a zombie basically’ [Ian, Interview 1, lines 8-10].
The term ‘straight-A student’ is an example of chained metonymy – first, it is the rating
of performance documented in a report card that stands for the quality of student, then
it is that performance as a student that stands for the self. Zombie is a popular
metaphor, the walking dead that populate screens both large and small, here invoked
to describe the experience of identity living with a mental illness.
Ian also described how a psychotic experience affected him particularly because of a
distinct attribute that formed his identity prior to becoming ill:
Chapter 5 Results
96
‘I did hear one voice only, ever in my, the whole experience of being in hospital,
and that [said] I was going to hell. I’m a Christian guy so to hear that voice
scared the life out of me, for the next ten years affected me’ [Ian, Interview 1,
lines 30-33].
In the previous example, he articulated the shift in his identity using global terms. In this
example, he reports the experience of psychosis as directly related to his identity, with
the power of the voice stemming from its exploitation of his belief system. This recalls
the ‘affront to yourself’ that Shirley described, though in this instance it is the actual
experience of psychosis rather than its diagnosis that leads to the effect. Ian
commented on the enduring emotional effect of the voice he heard, which persisted
long after his recognition that no external voice was involved, and that a plausible
explanation for the experience related to the internalised guilt he felt over his use of an
illicit substance prior to his first psychotic episode. In other words, the voice he heard
enacted a conceptual transfer giving meaning to his subsequent experience of
psychosis as a deserved punishment.
Ian voiced the strongest description of the influence of a diagnosis of mental illness on
his identity and its subsequent impact. He reported he had been diagnosed with
schizophrenia, and had multiple experiences of deterioration in his mental state, many
of them resulting in involuntary admission to hospital. He reported that these episodes
frequently occurred after he had ceased taking anti-psychotic medication, without
alerting his treating team. In the last few months, he had received a second psychiatric
opinion, and his diagnosis was provisionally revised to bipolar disorder, with a
differential diagnosis of schizoaffective disorder. He reported he was subsequently
adhering to the same anti-psychotic medication as previously prescribed, but now with
the addition of a mood stabilising agent:
‘Whereas before I was just totally disinterested in my condition, ‘cause I didn’t
want schizophrenia, I didn’t believe I had schizophrenia, the stigma that comes
with it, I was, you know, ashamed of for 20 years, so, but now, believing that I
have the right diagnosis, I’m happy to take medication for that’ [Ian, Interview 1,
lines 549-552].
As he reported it, his readiness to take the prescribed treatment rested not in his
experience of mental health or ill health, nor in the effects of the medication, but in the
name given to the condition for which he was prepared to accept treatment. This
demonstrates the conceptual power that resides in the metonymic shift, with the
Chapter 5 Results
97
change of name given to his condition governing his willingness to assume control of
his experience.
Ellen repeatedly used expressions that suggested a causative relationship between her
identity and her experience: ‘because I’m who I am’ [Ellen, Interview 1, line 29],
‘thankfully I am me’ [line 149], ‘cause I remembered who I was’ [line 151], ‘a lot of
things have happened to me as a result of me being who I am’ [line 269], ‘I refused to,
‘cause that’s not who I am’ [line 440]. Ellen’s identity stands for her values, even as the
specific values invoked are left underspecified. The repeated phrases are ostensibly
tautological, but rely on pragmatic inferencing for the selection of which unspoken
attribute is salient, be that courage, forbearance or resilience.
Other participants described different strategies for managing the shift in identity
caused by mental illness. Nathan played with the romanticised idea of returning to his
premorbid self, weaving through some of the more colourful metonymic transfers of
meaning from substance use to identity:
‘It just didn’t feel right, I didn’t feel like my old self, even though my old self was
a bit of a layabout, bit of a dope fiend, well not a dope fiend, a bong rat’ [Nathan
Interview 1, lines 174-175].
‘Layabout’ evinces metonymy whereby the (in)action comes to stand for the person,
while the attributes attached to substance use, both demonic and murine, are almost
comically negative. These metonymic references also suggest that these terms have
particular understandings within a microculture of language use, with certain concepts
being salient for participants in that group. For those outside, it may not be possible to
infer the distinctions between ‘dope fiend’ and ‘bong rat’, though for this speaker, the
shift in nuance was clear.
Nathan voiced the positive aspect of linking identity to a specific attribute:
‘There were some ups, yeah, my first gig with a show called XXXX, cause I’m a
roadie, my first gig at XXXX, that was a bit of an up, watching them onstage, as
a roadie, yeah, just anytime that I was working was an up’ [Nathan, Interview 1,
lines 205-207].
Here it was his work role that gave him a sense of value, the very act of watching the
band transformed by his experience of doing so from a specific and recognized identity
position: ‘as a roadie’.
Chapter 5 Results
98
For Ethan, who first developed schizophrenia in the 1970s, his reported strategy was
not to return to a prior identity, but to adopt an alternate identity by highlighting other
available attributes:
‘I was trying to mask my mental illness by being seen as a dropout or a hippy or
a drunk or a drug addict because there was nothing worse than, you know,
being seen as a lunatic, you know, because you had no credibility, you’re not
taken seriously, in those days people didn’t know how to treat people with a
mental illness’ [Ethan, Interview 1, lines 53-57].
His masking strategy involved adopting a series of other identities, all of which were
preferable to being perceived as mentally ill within an unwritten but nonetheless
powerful hierarchy of stigma. Questioned further about this in the second interview, he
laughed about his success:
‘I think it was a very unsuccessful strategy because people thought I was crazy
as well as drunk’ [Ethan, Interview 2, lines 54-55].
Each of these identities, dropout, hippy, drunk, drug addict, lunatic, relies on a
metonymic transfer of an attribute to speak the whole of his identity. Notably, the
selected attributes of the less stigmatized identities, lack of vocational functioning,
physical self-presentation, and substance use, all carry a potential to be voluntarily
relinquished. ’Lunatic’ however, carries the mark of permanence, and involuntary
attribution.
Ethan goes on to contextualise the alternative options that were available to him at the
time:
‘[In] the 70’s, there was a kind of, round, you know, in the dying days of the
Push, you know, there was a kind of rebellion and actual affirmation of
oppressed groups, I didn’t see it very much reflected in mental health stuff then,
but certainly with the women’s movement, the gay movement, student power,
anti-Vietnam war protests, the hippy movement, which is somewhat discredited
now, but it was very strong then, you know, there was multiple avenues for me
to try and disguise my mental illness as alienation or disaffection with the
system’ [Ethan, Interview 2, Lines 60-67].
Implicit in his description is the fact that a key element of the majority of the other
movements he names involved assuming identity positions related to membership of a
specific community.
Chapter 5 Results
99
Michael’s approach to distancing himself from an identity formed in relation to mental
illness was different again:
‘I mean it’s alright if like, 10% of the time I go, “Oh well, yeah, that’s me as a
schizophrenic”, but I don’t want, you know, 70% of me to be that schizophrenic’
[Michael, Interview 2, lines 188-190].
He expanded on this, explaining that, for him, it was the whole process of an attribute
standing for his identity that was questioned, not simply the choice of valued attributes
over non-valued ones. He reported the possibility of enjoying aspects of himself, while
retaining separation from globalising suppositions based on these enjoyments:
‘I would go to church occasionally but I wouldn’t be comfortable labelling myself
a Christian, yeah, I don’t really see much of a point in labelling myself, I know
who I am no matter what my activity is, yeah, like, I enjoy bushwalking but I
don’t call myself a bushwalker’ [Michael, Interview 2, lines 196-199].
In both instances, rejecting identity imposed or presumed on these other selected
attributes supported him when it came to maintaining identity separate from the
experience of mental illness, no matter how influential or traumatic that may have been.
Participants’ responses to the influence of the experience of mental illness on their
identities included elements of choice within the process, even though they did not
choose to have mental illness. Participants also described experiences that arose
subsequent to their diagnoses that they did not choose, specifically, the experience of
stigma.
STIGMA
All participants reported experiencing stigma related to their mental illness. This stigma
occurred from strangers, from friends and acquaintances, from family members, and
from mental health workers. Participants also reported experiencing stigma related to
attributes other than mental illness, including race or ethnicity, gender, employment
status and faith. A number of participants also reported self-stigma, that is, negative
thoughts about themselves related to having a mental illness.
Ethan identified the role of mass media, and the representations of people who live
with mental illness it reinforces, as contributing to the experience of stigma:
‘I also see that the media has a certain view of mental illness that’s different
from mine, even today there’s, there was a story about a mentally ill man who
Chapter 5 Results
100
might be somewhere in the US that killed five people, so big headlines ‘Mentally
ill’ so even though it’s probably true, he was mentally ill, it, just having it there
actually feeds that view of us you know as unstable, violent, it’s very difficult’
[Ethan, Interview 2, lines 273-278].
This is echoed by Yves:
‘You hear them say, you know, “Oh, some psych patient’s going to grab a gun
and kill people”’ [Yves, Interview 2, lines 171-172].
Yves’ experience of mental illness spanned decades, and he observed that stigma was
becoming more noticeable in recent times. Whereas in his early encounters with
strangers and police when he was first psychotic he was treated with kindness and
some humour, he now reported increasingly frequent use of derogatory language from
strangers and acquaintances, and more social withdrawal by others once his diagnosis
was known:
‘I think it, like, people, the way people talk about psychiatric patients and stuff
now, I hear more, I hear a lot of like, nonsense all the time, and I never used to
hear that, yeah, you know they call you loony or something, or you know, or a
psycho or something, and sort of like, they don’t have an education sort of, on
schizophrenia or bipolar or manic depression or anything’ [Yves, Interview 2,
lines 88-94].
Yves articulated a direct link between people using stigmatising language and people
having a lack of education about mental illness.
Ian used an idiomatic expression to describe the experience of stigma:
‘People have always got it in the back of their mind, that something’s wrong, or
not quite right’ [Ian, Interview 2, lines 116-118].
This idiom blends metaphor and metonymy; the mind is metaphorically described as a
container, and a thought, ‘that something’s wrong’, metonymically located spatially
within that container, specifically ‘in the back’, that is, not in plain view. Ian, like Yves,
drew a link between the lack of understanding of mental illness by the broader
community and its correlation with stigma. As he explained it, this ignorance was less
about how a person may or may not be expected to act in public, but about the actual
experience of mental illness:
Chapter 5 Results
101
‘The torment you go through, it’s the torment that you go through, that you need
relief from, and anguish, the pain that’s on the inside, I mean you can never
explain to someone, if someone sees a big gouge out of your arm at least they
can see the pain, they can sympathise with what pain you might be going
through, but mental illness, you can’t see anything, so no-one can sort of have
any sort of sympathy or empathy for you, really if they doesn’t, if they don’t
know, if they’re not a professional in the field, the general public have no idea
about mental illness and are quite scared of it’ [Ian, Interview 1, lines 608-615].
He suggested that the lack of understanding of mental illness in the community is
associated with fear of it, and of the people who experience it. For Ian, this ignorance is
based on the invisibility, and the ineffability, of the suffering experienced by people with
mental illness.
Ian also pondered how much of his own sense of stigma in relation to the diagnosis of
schizophrenia is related to the word itself:
‘I don’t like the word schizo for some reason, so schizo, even schizoaffective, I
don’t like the word schizo, I don’t know why that is, so there is a stigma even in
the word for me, because the whole idea of schizo is double-minded and
shattered, and all these sort of images, sort of something is, yeah, broken and
not whole, you know what I mean? And, just thinking about it now, schizo to me
sounds like a word that is shattered, like it’s a shattered word, does that make
sense? If you’re talking about your mind, I don’t like to think of your mind as
being shattered, and, you know what I’m saying? So I think I have a stigma
towards the word more than the illness’ [Ian, Interview 2, lines 391-399].
This passage demonstrates a process of thinking aloud; Ian articulated that he was
thinking through these ideas as he spoke. His analysis of his own responses shifted
back and forth between examining the etymology of the term ‘schizo’, and an
exploration of his reactions to the sound of the term, that lend it almost onomatopoeic
strength. He went on to ask a metalinguistic question:
‘You know, why don’t they call it a nice name? [laughing] “Ian’s a brainiac” and
then people would go, “Oh, OK”’ [laughing] [Ian, Interview 2, lines 435-436].
While this suggestion was offered light-heartedly, it is consistent with the particular
intensity of his responses to the sound of the prefix ‘schizo’. On this last idea about the
influence of naming, Shirley stated an opinion which stands in direct contrast with Ian’s:
Chapter 5 Results
102
‘I mean, I personally think stigma follows around whatever language you
choose for the experience’ [Shirley, Interview 2, lines 27-28].
She prefaced this statement with an observation that, despite the term ‘manic
depression’ being replaced by ‘bipolar disorder’ thirty years ago in clinical use, she still
encountered the term regularly from members of the community, frequently
accompanied by stigmatising attitudes.
Though Shirley’s statement contradicted Ian’s in relation to the influence of language
on stigma, her observations about the effect of the invisibility of mental illness echoed
his, and raised further issues:
‘I would classify what I experience as pretty invisible disability and hard to
articulate and describe and ask for accommodations around’ [Shirley, Interview
2, lines 183-185].
Like Ian, Shirley described mental illness, and the disability associated with it, as both
difficult to see, and difficult to articulate. As a result, she had experienced difficulty in
accessing services and study support to which she was entitled, due to failure on the
part of agency representatives to recognise her mental illness.
Shirley also talked about how many of the images on posters and other resources
designed to combat stigma against people with mental illness ironically contribute to it
by portraying normalised versions of what it is to have mental illness. She observed
that people are typically presented as being in groups, well dressed, clean and smiling.
The real experiences of many people with mental illness, including social isolation,
mental distress, and medication-induced weight gain, are effectively erased from these
representations. Shirley reported that people who experience these things do not
recognise themselves in these images, but instead see repetition of the normative
expectations that they frequently struggle against.
Shirley offered further observations on the consequences of the invisibility of mental
illness, based on her experience:
‘I’ve learnt lots of common assumptions people make, and they’re usually
wrong, the main one being, I’m not the sort of person that gets depressed, I’m a
different type of person, I don’t, I’m not that type of person, so people must
have this fixed idea of who gets depression and who doesn’t, and I don’t fit the
mould apparently, I think ‘cause I’m chatty and I can still be chatty, within
reason, I mean, there’s a certain level of depression where I can’t function at all,
Chapter 5 Results
103
but I can be, as long as I’m kind of able to get out of bed and put clothes on and
get myself out the door, I can usually talk, but I can be in pretty bad shape and
still talk, and I think people associate being able to talk, or hold a conversation,
or pretend to be OK, as passing’ [Shirley, Interview 1, lines 470-481].
She described situations in which she found herself, a person with serious mental
illness, being assessed by others as not meeting the ‘fixed idea’ of what kind of person
has mental illness. There are advantages and disadvantages to this. On the one hand,
it meant she was not subject to the stigma reserved for persons identified with mental
illness, and this had pragmatic consequences, including her capacity to retain her job.
On the other hand, her own experience was rendered invisible, and she was exposed
to the expression of prejudice, even though it was not directly applied to her.
Tanya was characteristically pragmatic about how to minimise the potential for stigma:
‘I don’t go out in public and, you know, put a pole up and raise the flag’
(laughing) [Tanya, Interview 1, lines 128-129].
Her statement underlined the relative invisibility of her mental illness, and some
measure of capacity to control visibility should she wish to.
EXPERIENCE OF SYMPTOMS
Participants spoke of the specific experiences of psychosis they had. These included
delusions, hallucinations and thought disorder. At times participants used these clinical
terms directly, but they also used the term ‘voices’ for auditory hallucinations. The
experience of psychosis was frequently reported as scary or disturbing, but not
universally so; several participants reported some enjoyment of their experiences.
Michael used simile to convey his experience:
‘One time I came home after class in my room, and all my classmates’ voices
were in my head as if, as if I was like at a function, with everyone talking around
me, [circling hands around back of head], and that was really scary and
frustrating because my own head was creating all this noise, it’s like going to
see a concert where there’s noise musicians just making a whole racket in my
head’ [Michael, Interview 1, lines 38-43].
In both cases, the terms he used evoke social situations where he would expect to
hear noise, a function and a concert. However, it was the coupling of this sense of
other people deliberately making noise with the recognition that it was his ‘own head
Chapter 5 Results
104
creating all this noise’, that gave the experience its emotional effects of fear and
frustration. Shirley and Ian described how difficult it was to articulate the experience of
mental illness. Michael, in his efforts to convey what it felt like to experience psychosis,
used figurative language.
Michael further grappled with how to accommodate the experience of auditory
hallucinations, identifying that the evidence of mental illness is inaudible for others, as
well as invisible:
‘I’m thinking of auditory hallucinations, I know people say, “Oh yeah, they’re
real”, but they might be, I don’t know, I don’t see that they are real, I mean it’s
OK for them to say that they’re real, just like a thought is real, but nobody else
hears the thought. You can’t record auditory hallucinations, put them on a tape,
that’s why I like tangible things because I can remember tangible things better’
[Michael, Interview 2, lines 372-376].
Michael identified that being able to hear something does not guarantee that it is real,
and pointed out that all thought, not just auditory hallucinations, cannot be heard by
others. He expressed his own preference for another sense, touch, ‘because I can
remember tangible things better.’ The last point introduces not just another mode of
perception, but also one of cognition, that is, memory, suggesting that it is not just the
event experienced in the moment, but also the way the event is recalled that
contributes to its impact.
Yves echoed Michael’s sense of recognition that voices aren’t real, but described how
nonetheless the experience interfered with his functional capacity:
‘I’ve never thought the voices have been real or something, yeah like the
delusions, but it does sort of interfere with your everyday life, showering even
sometimes, and going out and, you know, making sure you’ve got clean
clothes, not lying in bed, cause the voices would be so bad sometimes I’d just
have to lie in bed, just take it easy and that’ [Yves, Interview 1, lines 111-115].
Yves reported that his experience of symptoms was sometimes so bad that he could
not go out. This is another example of how aspects of mental illness can be invisible to
the broader public as, literally, the person remains unseen.
Yves described how his symptoms interfered with his functional capacity. Ethan
reported the disturbing experience of thought blocking; rather than a symptom getting
in the way of his thinking, his thinking itself was blocked:
Chapter 5 Results
105
‘When I went to university it was really developed in a big way where I had the
whole range of different things, one of the most unsettling was a kind of thought
blocking where I felt my brain was switching off and I couldn’t concentrate this
would mean I could look at, start reading a book and you know stare at the one
paragraph for a very long time and go into a kind of dream world, looking back it
was a bit like being asleep and having a dream or hallucination rather than
being in reality’ [Ethan, Interview 1, lines 16-21].
For Ethan, as for Michael, the experience and the memory of the experience were
linked. Again, like Michael, Ethan used simile to capture what it felt like, ‘a bit like being
asleep and having a dream’. Ethan also described how his thinking about how he might
recover was altered:
‘I had this insane theory that if I, you know, descended into madness and sort of
somehow came back in reverse the same way, I’d become well again, that I
could do it myself’ [Ethan, Interview 2, lines 265-267].
Ethan’s description of his theory is articulated in spatial terms, in particular, his use of
the phrase ‘descended into madness’. Ethan demonstrated use of a form of
predicational metonymy widespread in the community, transferring a property from a
human to a non-human object, mentioning his, ‘insane theory’. This usage also evinces
the prevalence of terms denoting mental illness being used loosely, even by people
with increased awareness of the influence of such language.
Ethan drew a direct link between some of his experience in the real world, and his
delusional thoughts. In particular, he reported a sense of powerlessness after he
became ill, that had material effects on his life in terms of the disruption to the
vocational trajectory he had previously been on. This powerlessness periodically
resurfaced in the delusional belief that he was a political prisoner, which in turn offered
an explanation for the shift in fortunes he had undergone.
Shirley also reported experiencing this type of delusion. In two episodes of acute
psychosis which occurred fifteen years apart, she was listed as a missing person, as
she left where she was staying without informing anyone. In both cases, she had the
belief that someone intended to do harm to people close to her, specifically because of
her actions. In the second instance, her belief was that drug companies were planning
to abduct her relatives to punish her for her advocacy work around side effects. It is
possible to trace a metonymic pathway for these delusional beliefs.
Chapter 5 Results
106
Shirley pointed out that not all of the symptoms she experienced were unpleasant,
including this wry account of the influence of environment on the experience of
psychosis:
‘London’s quite a good place to sort of flip out, because everything actually is
sort of connected, ‘cause there’s all, you know, every building’s designed by
Christopher Wren, or there’s some, there’s always coincidences everywhere,
you know, or you’re standing on someone’s grave when you don’t realise it on
the footpath, and you know, so it’s actually pretty trippy’ [Shirley, Interview 1,
lines 210-214].
Her description captured the way in which meaning can be attached to place, and how
this meaning can escape attention altogether or alternately be inferred correctly. Her
inference is that this kind of meaning is always available in certain contexts, and that a
shift in how one is mentally experiencing the environment can increase access to
inferring these meanings. This may give rise to another reason to ‘descend into
madness’ as Ethan phrased it. Shirley reported on the next phase of her symptoms:
‘I always liked to go and moon around Greenwich and, you know, hang out
round the Observatory Hill, and so I’d do a bit of that, and I remember coming
back by ferry from Greenwich one day and it felt like everything that was
happening in, the whole city was happening because of me, like it was a great
big procession and it was all caused, in and around and because of me, which
is quite extraordinary’ [Shirley, Interview 1, lines 265-269].
Her own assessment of the experience was that it was ‘extraordinary’, that is, the
inference that everything in the city was occurring because of her was not disturbing or
frightening. Her description, 'moon around Greenwich’ deftly blended a verbal phrase
metaphorically derived from cosmology with the location of an astronomic observatory.
In counterpoint to the problems identified with the invisible aspects of mental illness,
Shirley pointed out a very specific type of visibility she had in this phase of her illness:
‘When I’m like that I can be quite compelling and quite persuasive and quite
charming and quite mischievous, and quite energising to other people’ [Shirley,
Interview 1, lines 248-250].
The effect of her presentation at this time was to disguise the severity of her illness,
which deflected the attention of those trying to help her, deferred clinical intervention
and contributed to a dangerous deterioration in her mental state. This too was a matter
Chapter 5 Results
107
of pragmatic inferencing, with the attributes that were deemed salient by those around
her being read as positive aspects of her personality, leading them to incorrectly
underestimate how unwell she was.
Ian also reported ambivalence about his experience of symptoms:
‘I’d hate to stay like this, ‘cause I can’t sleep and I can’t settle, I can’t turn off,
but there’s part of it, the productive side of it, I definitely like, because normally
by nature I’m a bit lethargic’ [Ian, Interview 1, lines 318-320].
Ian’s reported experience of hypomania echoes Nathan’s evaluation of himself prior to
the experience of psychosis, both recognising patterns of low energy and motivation
before becoming unwell. Ian also related an experience of acute psychosis that he did
not find disturbing:
‘But the funniest story I could tell you about being manic and what I thought I
was going to be, become, I’ll tell you this, I thought I was going to become a
horse in the Melbourne Cup, like, I told this story in the foyer of the XXXX clinic
the other day, and like, what would happen was, I’d be sitting watching the
television, and then all of a sudden, just before the race started, I’d become the
horse, the favourite, like, say it’s Kingston Town, I’d become the favourite of the
race, but I’d know that I was Ian, as a horse, as the favourite, if that makes
sense, and so I’d just run, knowing I was going to win, and then I’d win the
Melbourne Cup, this guy that had just been in there for depression, balling his
eyes out ‘cause he was sad, turns to me and he goes, “Would you be a
gelding?”, and so he starts joking with me and in the end he was laughing his
head off, and that’s why I want to make my story known, because some of the
stories I have are quite funny, there’s stories I have that are sad, like the arm
and that, trying to cut off the hand, but there’s some stories I can tell you about
the manic side that are brilliant and make people laugh, and that’s why I like to
tell my story, because it can help people know that you can go from this to this,
and then back to normal, anyway, it’s a good story, and it’s just like, oh yeah, I’d
say, “No I’m not a gelding, the boys would be there, you know, they’d be there”,
and I’d pin the ears back and take off, and I’d, I’d win the Melbourne Cup, and
then I would, five minutes after the race I’d become Ian again, back on the
couch in his lounge room, having to go and have a shower because I’d just won
the Melbourne Cup as a horse [laughing]. That’s a good thought, isn’t it?’ [Ian,
Interview 1, lines 445-466].
Chapter 5 Results
108
Ian narrated his experience of novel symptoms. His stated evaluation was that it was a
good story, not a negative one. Moreover, he reported the retelling of the story as
performing a valuable function; practically through raising the spirits of a depressed
person he met in a clinic waiting room, and potentially through informing a wider
audience of how bipolar disorder can be experienced.
Ian’s narrative demonstrated a number of linguistic and narrative features. Within the
overarching narrative of his experience of being a winning horse in the Melbourne Cup,
he embedded a secondary narrative about another time when he told this story. The
way in which he was able to swap back and forth between the narratives, placing the
joking response to his listener some distance away from the reported question, and
interspersing the narratives with commentary on the power of stories, came across as
skilled storytelling, rather than disorganised thinking conveyed via circumstantiality.
His retort to the question ‘Would you be a gelding?’ played with use of a whole for part
metonymy, ‘the boys’ standing for the body part excised from a gelding. He conveyed
his winning speed through an idiom, ‘pin the ears back’, which again features a bodily
metonymy, with ‘ears back’ standing for the overall streamlined shape necessary to
win. The intensity with which his experience of the winning the race as a horse was
embodied was strikingly evinced by his need to shower afterwards. It is, of course,
possible that he was sweating with excitement on a balmy Spring afternoon watching
the race, and that this ordinary bodily experience lead to his delusion rather than vice
versa, but this is not the meaning that he inferred from the experience.
Nathan reported a transient but powerful belief that he was going to die when he was in
hospital, because he had been placed in a room near older people. He also reported
that he believed every passing voice he heard from the corridor was talking about him.
This latter experience could be described clinically as an illusion, as the meaning he
inferred was prompted by a real, and not imagined, stimulus.
Not all participants reported interest in the meanings inherent within their psychotic
experiences. Tanya dismissed the idea that the content of her auditory hallucinations
may have held meaning: ‘Oh, it was just all this racist crap’ [IV 1 line 176]. She reported
the content of the hallucinations was not dissimilar to language she experienced in
reality, and she dismissed both as not being important to her.
Chapter 5 Results
109
EXPERIENCE OF TREATMENT
Participants described a range of experiences of treatment, both positive and negative.
POSITIVE EXPERIENCES
Participants spoke of engagement with mental health workers as key elements in their
experience. Matthew was broad in his praise:
‘I must admit, I have very fond, fond memories of all the nurses and doctors at
both XXXX hospital and XXXX clinic, and I have nothing but the greatest
admiration for them, they really helped me’ [Matthew, Interview 2, lines 118120].
This praise was repeated by Ellen, whose evaluation introduced a further element, the
idea of being treated as a human being, in a relationship with other human beings:
‘I’ve been blessed with having nurses who actually have treated me like a human
being, and it’s just a natural kindness and respect, irrespective that we happen to
be in a psych ward and they’re in the role of the nurse’ [Ellen, Interview 2, lines
30-33].
Ellen identified both ‘the role of the nurse’ and the environment of the ‘psych ward’ as
potentially inhibiting the ‘natural’ occurrence of this interaction between human beings
by highlighting that the interactions took place ‘irrespective’ of these factors. Ethan
echoed the importance of being treated ‘as a human being’:
‘That thing about key people being there and treating me as a human being, not
a circus freak or something like that is really important to me’ [Ethan, Interview
1, lines 211-213].
He also used a metaphor ‘not a circus freak’, invoking the spectre of a stigmatised
category of persons from a previous generation. Again, like Ellen, Ethan highlighted the
possibility that he may not have been treated as a human being when engaging with
mental health workers, even while reporting that he was treated well.
Ethan also talked specifically about the careful use of language by his treating
psychiatrist:
‘He actually was very good and gentle in explaining the fact that because of
this, he thought that schizophrenia was a label, and that it, sort of, broadly
speaking it fitted me, but he wasn’t going to say “You’re schizophrenic”, you
Chapter 5 Results
110
know, he was just going to actually work with that’ [Ethan, Interview 2, lines
171-174].
This therapeutic encounter occurred in the 1970s. Ethan reported his psychiatrist
stating that the diagnosis could function as a label, and explicitly making clear that he
would not use this as a global term for him, but rather use the term while keeping it
separate from Ethan as a person. Ethan was the only participant who had not been
treated as an involuntary patient under mental health legislation, and he attributed this
outcome to a combination of factors, including his longstanding collaboration with
mental health workers.
Medication
While many of the positive experiences of treatment reported by participants were in
relation to their engagements with mental health workers, seven participants also
explicitly stated that taking anti-psychotic or mood-stabilising medication significantly
contributed to their mental health. Most of these participants reported that getting the
right medication, often after long periods of ineffective pharmacological treatment, was
a key step in their recovery.
For Matthew, who reported he had recently remained out of a psychiatric inpatient unit
for twelve years, after multiple involuntary admissions between the ages of 16 and 37,
medication was one of three critical factors in his recovery:
‘What changed was, a more inner understanding of myself, my religious faith,
and the very good and the very important help of the medication’ [Matthew,
Interview 1, lines 50-51].
Yves reported waiting a long time before finding the right medication:
‘I probably heard voices for about 20 years before I found a really good
medication that took them away’ [Yves, Interview 1, lines 30-31].
He had also experienced many involuntary admissions over the previous 20 years, but
at the time of interview he had not been in hospital for three years. The benefit Yves
enjoyed from this change went beyond avoiding coercive treatment, as the severe
effect of the voices on his capacity to function was noted previously. At the time of the
interviews, Yves had also successfully made the move from living in an institutionalised
boarding house, to living in independent housing.
Chapter 5 Results
111
Michael also traced a broader change in his circumstances as a result of finding the
right medication:
‘It’s my autobiography and thankfully clozapine has changed that story a bit,
and yeah, like I’m working in mental health now, that’s a big change from being
just a consumer’ [Michael, Interview 2, lines 218-220].
He drew a direct link between the effects of the medication on his vocational capacity,
and on the narrative of his life.
Shirley also noted the impact of a particular medication on her functional and
vocational capacity:
‘It wasn’t really much of a life and for me what ended up being quite effective
was one of the epilepsy drugs which is lamotrigine, which as far as I know is still
not on the PBS [Pharmaceutical Benefits Scheme], and that’s something that
really needs looking at, because to be quite frank, that has been the difference
between me struggling to type numbers in a data entry screen, and being able
to work professionally as a trainer and travel and do honours degrees and, you
know, so I think I, you know, I think there’s public policy things that need to be
considered when we look at having the full suite of options open’ [Shirley,
Interview 1, lines 547-554].
Shirley drew on her own experience of achieving significant benefits from a specific
medication to state an advocacy position in relation to ensuring access to the most
effective treatments for everyone affected by mental illness.
NEGATIVE EXPERIENCES
All ten participants reported negative experiences in relation to their mental health
treatment. These reports were offered spontaneously, as part of their responses to the
request to narrate their own experiences. The events they reported as negative
occurred across the time span of participants’ experiences, from the 1970s until the
time of the interviews, 2014.
Ellen reported that in her experience, the attitudes that mental health workers
demonstrated through their language were problematic. Her observation was that they
seemed to link mental illness with lack of intelligence:
Chapter 5 Results
112
‘The language that’s used is very much, in my experience, is you’re spoken to
as if you don’t have a mind, as if, as if there’s no intelligence and you don’t
understand’ [Ellen, Interview 2, lines 21-23].
For Ellen, this was not restricted to language, ‘With the language goes the tone and the
attitude.’ [Ellen, Interview 2, line 40], and incorporates non-verbal communication
including gestures, ‘It’s very difficult to communicate with someone when they’re
looking at their watch.’ [Ellen, Interview 2, lines 208-209]. This last observation
depends on a non-verbal metonymic inference, that a worker looking at their watch is
signalling impatience by drawing attention to the time. Later, Ellen expanded on her
perception of a further aspect of workers attitudes, that a person under an involuntary
treatment order lacks capacity to make decisions, and therefore must have actions
taken on their behalf:
‘The mental health professionals in the hospital, yep, it would, they, they just
had, you just were treated differently, from the moment you give signed and,
you know, you’re scheduled, that’s it, you have no, you have, you can’t think for
yourself, so we’re here to think for you, that’s pretty much what happens’ [Ellen,
Interview 2, lines 341-345].
Ellen told her story in her own style, which was characterised by these kinds of
assertions about other people’s thoughts. In narrative terms this is an example of
argument, as she adopted a position from the present perspective on past events. Her
viewpoint was expressed, but not followed by concrete examples that would illustrate
her statements. Ellen’s speech also frequently demonstrated the kind of shifts evident
in this example, with sentences and even parts of sentences left unfinished before she
moved on to the next point.
Ian reported his experience as a mental health inpatient in the 1990s in graphic terms.
He stated that while he was dealing with overwhelming guilt, depression and
psychosis, he had absconded from the ward, and tried to cut his wrists. Inspired by his
belief in Jesus, he then changed his mind and returned himself to the ward:
‘I went back there, got put in a room by myself for half an hour before the doctor
arrived, wasn’t offered a drink, I was absolutely gasping for water or something,
doctor came in, stitched me up, there was no-one, no-one available to counsel
me, ask me where I’d been for five hours, or you know, no-one asked me any
questions, they just left me and took me down to, after I had the stitches, took
me down to ward XX and threw me in there like I was being thrown into gaol,
Chapter 5 Results
113
and then at night, the nursing staff were asleep, so all the other patients would
come and steal your bedding while you were trying to sleep, it was just
atrocious, and you’re in absolute agony with your arm, and all this is going on’
[Ian, Interview 1, lines 278-286].
Ian reported that his visible physical injury was treated, while other, less visible,
aspects of his physical well-being, including thirst and pain, were not dealt with by the
staff. Moreover, he reported that his mental state, which had led to his self-injury, was
not addressed.
Ian was also critical of the inferences he perceived his treating team made about his
response to electro-convulsive therapy (ECT):
‘When I started ECT, it takes away your memory, so I forgot what I was worried
about for a while [laughing], so it looked like I was getting better, but they
weren’t actually dealing with the problem’ [Ian, Interview 1, lines 183-185].
For Ian, it was clear that the inference that the treatment was effective was incorrect,
and he reported he was discharged still suffering from the depression with which he
initially presented, but with the additional problem of acquired memory loss from ECT.
Ian expressed some pessimism about his experiences with inpatient services, and the
inferences that he saw in effect, summarised by his stated opinion, 'From a health
worker point of view, you’ll always be mentally ill.' [Ian, Interview 2, line 124]
Nathan was also critical of the treatment he received after a suicide attempt:
‘I was put on suicide watch all night, or what I thought was suicide watch, it was
just a security guard sitting outside the door the whole night’ [Nathan, Interview
1, lines 192-193].
He identified a gap between the purpose of the protective intervention as stated in the
language, and its implementation in practice. Nathan also reported how he used
language to influence his treatment:
‘I remember going in to see a doctor and we talked about what was the best
plan for me to keep well, and I just spouted all this bullshit just to get out of
there, because I just was scared of getting admitted’ [Nathan, Interview 1, lines
78-80].
In his telling, he implied that the doctor colluded with him in ostensibly respecting his
version of insight, in order to expedite his discharge from the hospital.
Chapter 5 Results
114
Other participants reported their experiences of gaps between service policy and
service delivery. Yves reported times when he had been refused anti-cholinergic
treatment to reverse the side-effects of anti-psychotic medication, even when clinicians
acknowledged that he was clearly exhibiting signs of dyskinesia. His stated view was
pessimistic:
‘You don’t have any say in it, it’s sort of their system’ [Yves, Interview 1, lines
234-235].
Yves identified that the ‘system’ was enacted by individual clinicians and in his view, at
times it was operating to meet their needs rather than those of the people receiving
care, even when this meant foregrounding a history of anti-cholinergic abuse over
present clinical evidence in a treatment decision.
Oliver was critical of his treatment as an involuntary patient of public mental health
services, and he also invokes the system as an important element. The following
lengthy excerpt from his narrative demonstrates the language practices he used to
express his thoughts:
‘It’s sort of like, what we could afford and stuff, and sort of, how it was because I
think maybe if I went to like, XXXX [private clinic] or something, and was just
under supervision for a while and had counselling and stuff, but because I go to
XXXX [public hospital] and everything’s happening so fast, and they want to
straight away diagnose and medicate and, sort of, get you out of there, yeah I
think, yeah, my parents didn’t know what to do, so they just listened to, “OK,
that’s what’s meant to happen”, so yeah. I did feel anger for quite a while but,
like, not angry at any one particular person, because I think every psychologist,
psychiatrist I spoke to and that, they seemed to have my best interests at heart,
they didn’t seem like evil people or whatever, but yeah, just maybe the system a
bit but, yeah it always comes down to, to the facilities, if it’s, if there’s money I
guess, if there’s enough money for everyone to get the proper treatment.
[pause]. What happened, when I went there, that got me angry, because I
spoke to the doctor when I arrived and I told them about this, I think I can, you
know, you know, and she said, “OK, we’ll put you in under supervision”, so I
went in and I said, “OK, I’ll go in under supervision”, and next morning I woke
up and the nurse came and tried to give me this medication and I said, “Oh,
they told me I’m just under supervision”, I said, “Who authorised that?” and they
said, “Oh a doctor this morning”, and I said, “Oh, but I haven’t seen a doctor
Chapter 5 Results
115
this morning”, and I said, “So how can someone authorise something without
even meeting me?”, and they said, “Oh, they read your file”, and I said, “I’m not
a piece of paper”, I said, “I’d like to talk to the doctor please”, and I spoke to the
doctor and they said, “OK, we’ll have you under supervision”, and I didn’t really,
for two weeks I was OK in there, like I was never aggressive to anyone or
whatever, maybe my thoughts, yeah I thought I was OK, but yeah, yeah [pause]
I think with communication, [it] happened with the nurses a few times and that,
they sometimes, they don’t know, but they just make something up a bit, like,
and they should be just honest, like, even if it’s hard to hear and the person
might react in a bad way, they should still get the truth before, like anything
else, you know, because if you start to not trust the people who are there to
help you, then you sort of, yeah [long final pause]’ [Oliver, Interview 1, lines 4777].
In the first section of the passage [lines 47-58], he stated a general critique of the
public mental health system. The language he used here was relatively vague; the
passage includes fillers, ellipses, and metonymies. The fillers ‘and stuff’ and ‘sort of’
are each repeated more than once. He reported his parents’ confused reaction to the
speed with which events were occurring using ellipsis, ‘they just listened to’ [line 52]
and projected resignation ‘OK, that’s what’s meant to happen’ [lines 52-53]. He stated
his anger at his treatment, but deliberately deflected this from individual health workers,
instead metonymically highlighting ‘the system’ [line 56], and ‘the facilities’ [line 57]
before suggesting an argument based in health and economic policy, ‘if there’s enough
money for everyone to get the proper treatment’ [lines 57-58]. This passage is
categorised in narrative terms as a report.
After a pause, Oliver resumed speaking, but he now narrated a specific episode that
made him angry, where an agreement that had been made with him to remain in
hospital under observation was changed without any consultation [lines 58-69]. He
described in close detail a series of verbal exchanges between himself and various
mental health inpatient staff over one night and one morning. His language changed
markedly in this passage. He repeatedly used direct speech, with the phrases ‘I said’
and ‘they said’ followed by words reported as direct quotations, ostensibly reproducing
these interactions verbatim. He again made strategic use of metonymic language. In
response to the reason he was reportedly given for the lack of consultation ‘“Oh, they
read your file” [lines 66-67], he made explicit that the consideration of a document
about him had been made to stand for direct communication with him. In order to
Chapter 5 Results
116
assert his right to have a decision regarding his treatment appropriately discussed with
him, he deliberately extended the metonymy ‘file’ used by staff further, and stated, ‘I’m
not a piece of paper” [line 67], creating a metonymic chain to underline his point. Oliver
reported that he was subsequently seen by the doctor, and the initial treatment
agreement upheld.
Following this narrative of a discrete episode in his care, Oliver shifted once more to a
different narrative mode, in which he evaluated his experience of communication with
mental health workers, particularly nurses [lines 71-77]. His language again became
more general, with repeated reference to a non-specific ‘they’, again deflecting from
ascribing individual responsibility. He suggested that nurses avoided clear
communications from a fear that, ‘the person might react in a bad way’ [line 75]. He
reported the belief that this avoidance on the part of nurses could lead a person to lack
trust in their treating team, but in the end he elided stating explicitly what the result of
that lack of trust may be. Oliver opted not to participate in a second interview, as he
reported the first interview raised unwelcome memories. I was therefore unable to seek
clarification with him about the possible motivation for his language choices.
Shirley highlighted the way in which interactions with the mental health system can
compound the trauma of experiencing mental illness:
‘Psychosis itself can be a traumatic experience, the actual scary thoughts, and
then on top of that, service system related trauma, you know, the things people
have said, or the time they didn’t help or the time you’ve been in seclusion’
[Shirley, Interview 1, lines 818-821].
In her description of trauma that occurs in relation to the system, the first item she
noted was what mental health workers said, that is, the trauma is caused by language.
The trauma of not being helped echoes Ian’s experience noted above.
Shirley illustrated an instance where she did not receive help that made her angry
rather than traumatised. She had participated in the development of a resource to
guide improvement in health practice, specifically through psychiatrists assuming
responsibility for the monitoring and management of the long-term physical side-effects
of anti-psychotic medication among the people they treated. Shirley reported her own
treating psychiatrist displayed the brochure detailing the resource in the waiting room of
the clinic, but then refused to implement the practice in Shirley’s own care, saying ‘I
don’t have time’. Shirley’s view was that her psychiatrist used the time constraint to
mask an unwillingness to deal with the issue on a broader level:
Chapter 5 Results
117
‘The “I don’t have time” is just a “No, I don’t want to deal with that, because if I
deal with that with you then I’ll have to deal with that with all the other people
that I prescribe these things to”’ [Shirley, Interview 1, lines 745-747].
Shirley reported that, notwithstanding her prominent advocacy role and her anger at
the public support for the intervention not being matched by actual practice, she felt
personally unable to raise the issue or repeat the request to her psychiatrist.
Shirley also told of her dismay when she attended a conference on bipolar disorder,
where the depressive phase of the illness seemed to be presented as an emerging
topic for clinical attention:
‘I know there’s quite a lot of variation in how it plays out for people but, you
know, the bits that are actually ruining your marriage or stuffing up your career
or making your life a living hell is depression, and I just thought, the whole
profession has been totally focused on the stuff when we’re out of control and
turning up to emergency wards and bothering the public and getting brought in
by police, and that kind of private suffering hadn’t been kind of prioritised or
noticed on a kind of profession-wide scale, and I just was really astounded by
that’ [Shirley, Interview 1, lines 645-652].
This was another example of a participant noting the invisibility of aspects of mental
illness, though in this instance, it was the professionals whose role is to understand
mental illness who were perceived to have missed the obvious. Shirley explicitly drew
the contrast with the more visible aspects of the illness, including, ‘bothering the public’,
as a possible explanation for the collective myopia. A characteristic of Shirley’s overall
narrative was the capacity to link her own experience to broader systemic issues. This
was consistent with her longstanding role as an advocate.
Shirley reported an experience of sub-optimal health treatment for a physical problem
when her mental health diagnosis was raised in an emergency department:
‘The other types of situations where the language is really important and scary
is that issue of diagnostic overshadowing, and I turned up to emergency two
years ago and it was the first time in my life that I’ve ever had sudden breathing
problems, I’ve no history of asthma or anything…it didn’t go well, you know, I
said, “Well yes, I’ve got bipolar,” and then they just, what do you call it, the
triage nurse says, “Oh well”, I said, “But I’m having trouble breathing”, and she,
Chapter 5 Results
118
she said, “Are you delusional?”’ [Shirley, Interview 2, lines 333-336 and 347350].
Shirley also reported a negative experience with the health research system, when she
sought to undertake an exploration of the influence of cultural understanding on the
experience of being diagnosed with a mental illness, and was told that the particular
university department where she applied was only interested in the mental health
effects of diagnoses of physical illness. She presented a brief version of her analysis:
‘You know, people get given a diagnosis of schizophrenia and then they have,
like, quite a high risk of committing suicide in the several weeks after hospital,
and I think personally, a lot of that’s around, you know, you’ve just been given
the diagnosis of doom, you know, which, I’m not saying it is the diagnosis of
doom, but you know, culturally it is, and historically even the way medical
professionals talk about it and, you know, the prognosis for it and the way it
gets delivered, and you know, your life outcome assumptions are really poor,
and not that the actual outcomes are like that, so people get given this
horrendous kind of name, word, and there’s been kind of no academic space
within health psychology, as far as I can, well certainly not when I went looking,
to sort of say, “Well, what does it mean to get given this word schizophrenia,
and how does that affect someone?”’ [Shirley, Interview 2, lines 519-529].
Shirley identified that people who have the experience of being diagnosed with
schizophrenia already have cultural understandings of the illness that impact on their
capacity to process the information, as these are often reductive and negative. Her
stated view was that the way that this information is delivered is worthy of attention,
clinically and academically.
Clinical reticence
Participants reported a special type of negative experience of language used by mental
health professionals, the withholding of information about their diagnosis. This was a
form of clinical reticence. On the one hand, five of the participants directly reported the
experience of feeling that information was withheld from them. On the other hand, each
of them was forced to question this experience, and they each came to different
rationalisations for the practice. Participants expressed ambivalence about this
experience. Oliver’s analysis of this communication problem, where he felt that
clinicians, particularly nurses, were avoiding explicitly talking about his diagnosis and
Chapter 5 Results
119
treatment, was presented in the previous section, and he identified that a possible
outcome was that the person feels a lack of trust in the treating team.
Nathan stated, ‘It was taking a while to actually, well not to diagnose it, but for [them] to
actually tell me.’ [Nathan, Interview 1, lines 260-261]. He expanded further on this, and
how he perceived this use of language:
‘It’s always been, always the language that they use is very soft and placating,
which to me is, I sort of feel it’s almost to the point of condescension, where
they don’t really go through it in depth and say, here are the causes, here are
the effects, this is this, this is that, so, that’s been my experience’ [Nathan,
Interview 2, lines 77-80].
Nathan expressed frustration, both with what he perceived to be the delay in informing
him of his diagnosis, and the imprecise and superficial language in which information
was conveyed by mental health workers. This echoes Oliver’s suggestion that health
workers don’t always know what they are talking about. Nathan experienced this as
being treated with condescension. When asked how he himself described his condition,
Nathan responded, ‘a depressive disorder with psychotic features’ [Nathan, Interview 1,
line 256], that is, using clinical diagnostic terms.
Michael reported that, 'It never was talked about in a way that I could understand its
presence in my life’ [Michael, Interview 1, line 189]. While he noted this, Michael did
register that he lacked insight into his mental illness for many years, and it may have
been this that impeded effective communication between him and mental health
workers.
Ethan reported, ‘I had a kind of feeling that no-one seemed to tell me what the problem
was.' [Ethan, Interview 1, line 72]. Like Nathan, he used the language of ‘feeling’ to
convey his impression. His suggestion was that this was enacted by others as a
protective strategy:
‘I’ve thought, perhaps the doctors and family thought that I might get so upset
by the label that I’d try and kill myself, you know, I think that it might have been
a protective thing, I have no evidence of that but, you know, the way people
think to try and work out things’ [Ethan, Interview 1, lines 126-129].
This is a further example of the process by which people make inferences about events
in order to understand the meaning of an experience. It aligns with earlier examples
where participants have reported on reconstructing their memories rather than simply
Chapter 5 Results
120
recalling events. In this context Ethan reported he was trying to piece together the
meaning of other people’s actions, without clear evidence about the intentions behind
the communications of his doctors and family. Ethan also reported very careful
communication by one psychiatrist (in ‘Positive experiences’ section above), which
serves as a reminder that people can have very different experiences in their
interactions with mental health workers over the years.
INSIGHT
Insight is a key term in mental health discourse. Several participants discussed their
experience of awareness that something was wrong, but lack of knowledge about what
that was. For Yves, this period lasted several months:
‘You know I probably have spent two or three months before my first onset not
realising, you know, what I had or anything Andrew, and something was wrong
yeah, something was wrong’ [Yves, Interview 1, lines 397-399].
This suggested the complexity of using common words with particular inferences. Yves
expressed insight that ‘something was wrong’, at this point, however, he did not
express insight into the fact that what he was experiencing was mental illness. Ian
echoed this experience, also noting that when he sought assistance from others, he got
no further:
‘Oh, I knew something was wrong, but I could never explain what happened to
anyone, it seemed like no-one, any time I explained the situation, no-one
showed any sort of understanding as to what had happened’ [Ian, Interview 1,
lines 112-114].
These accounts demonstrated that, prior to experiencing mental illness themselves,
many people have minimal understanding of the issue, and that this lack of
understanding is broadly shared, for instance, in their existing support networks. These
ideas complement, rather than contradict, Shirley’s observation that a cultural
understanding of schizophrenia is that it is the ‘diagnosis of doom’. Both positions
evince lack of understanding of the reality of mental illness.
Michael reported an extreme relationship to insight into mental illness, in the opening
lines of the first interview:
I actually had the diagnosis before I had the realisation that there was a
problem, so, what led me to realise that I did have a mental health issue was,
again, being forcibly taken to a mental institution, a psychiatric ward, but this
Chapter 5 Results
121
time, and that had occurred, ten, ten or so times previously, but this time around
I was just, I was, I was really wondering, “What, what was putting me in this
situation?” I don’t know, I just didn’t have the sense previously that there was
anything wrong, it was just something I was going through, I didn’t really, wasn’t
really aware that there was reasons why I was being put into a mental
institution, so, I enter this ward at XXXX hospital and, I’m just puzzled by what’s
around me, not certain why it was happening, and I was starting to question it
myself, ‘What am I doing here, what’s going on?’, and during that stay in
hospital I kind of, it’s like the realisation dawned on me, with the help of a
psychiatric doctor whom I’d asked earnestly, ‘What do I need to do about this
situation that I’m in?’ and he told me that I needed to gain insight into my
condition, and I had, I had written in my diary previously that people had said,
‘You need insight’, but it never clicked, and now with this kind of questioning
attitude, puzzled by what’s going on, the doctor telling me that I needed to find
insight into a condition was proof that I had a condition, and that was the
moment when I accepted it myself, that this was not something that doctors in a
system of the medical profession was imposing on me, but there was a reason
for my, you know, for this past history and present of hospitalisation, and
dealing with case workers, and a whole lot of other staff, yeah, so that was that
was that, and that started me on a journey of trying to understand mental illness
and trying to live a life that incorporated acceptance of mental illness, but tried
to reduce the effects of the mental illness on my health’ [Michael, Interview 1,
lines 8-30].
In Michael’s account, he was able to sustain the lack of insight into his mental illness
through ten involuntary admissions to psychiatric inpatient units. He noted that he had
previously engaged in explicit discussions of the term, and had even recorded it in a
journal. Michael identified that the turning point depended on a combination of a mental
health worker reiterating the term, and his own willingness to understand the concept.
Without the latter, he had remained impervious to the idea, but it still required a
psychiatrist to use the term again for it to be available for him to process, the naming of
the need for insight into a condition standing as proof of the condition.
Michael went on to articulate that his own experience of developing insight contributed
to a sense of responsibility to help others:
‘I probably have a responsibility, having been through it and now taking
medication that has stabilised me, that I should try to assist and advocate for
Chapter 5 Results
122
other mental health clients, try and give them some insight hopefully’ [Michael,
Interview 1, lines 205-208].
As the term insight was raised spontaneously by several participants during the first
interview, I decided to ask everyone if the term had meaning for them in the second
interview. Thus, the following responses were prompted rather than spontaneous.
Ellen stated a negative view of the word as it is used in mental health practice:
‘See that’s funny I forgot, I hate that, the lack of insight, that’s thrown in my face
when I’m in there’ [Ellen, Interview 2, lines 473-475].
Her choice of words was idiomatic: ‘thrown in my face’ combines a metonymy, where
face stands for person, and a metaphoric reification, where a concept is given form as
a material thing that can be used with physical force. This is an example where the
complexity of everyday language can pass unnoticed.
Ethan initially used a metaphor to describe insight:
‘Insight is a double-edged sword I think, because if you allow it, you can
develop insight and become really depressed, because, you know, it’s one of
those things and insight for one person is not insight for another person, so to
me it’s about testing hypotheses, so if you, if I think, for example, if I think I’m a
political prisoner or, you know, people are out to get me or whatever, I really
make an effort to rationalise it. With insight, as in knowledge of medication,
knowledge of what schizophrenia is, I think that’s really important for us to be
able to ask questions of clinicians and actually get the best possible treatment’
[Ethan, Interview 2, lines 251-257].
‘Double-edged sword’ is a metaphor that infers a tool that can be useful and powerful,
but also dangerous for the wielder. Ethan set out the reasons for each possibility. On
the one hand, developing insight into having a mental illness can actually lead to
depression. On the other hand, it can be helpful. Ethan further identified two ways in
which insight can be helpful: on a personal level, it can function as a mechanism for
challenging paranoid thoughts as they arise; and on a clinical level, it can provide the
person with knowledge that contributes to shared decision making with mental health
workers, with a view to achieving optimal treatment.
Tanya’s response to the question about insight was a clear example of tangentiality;
she did not provide a direct answer to the question, nor was her response completely
Chapter 5 Results
123
without a link to the question, as the element she chose to expand on drew on the
concept of sight:
‘Insight, is that when you picture things in your head? I wouldn’t, I don’t know,
I’ve pictured a lot of things, but most of it’s been like, the sky when it turns
different colours, or fish at the pond like, cause I was feeding the eel and that,
and can picture the eel sometimes, swimming up and around, going up and
down like a dolphin, yeah, I think because I just, kept to myself so much
throughout these years and nobody, I mean insight, when like, nobody really
recognised me, because I was so quiet, and yeah, I just stopped talking for
about five years, and I just started like talking now, like communicating properly
and stuff with people, just like, last year or so’ [Tanya, Interview 2, lines 149157].
Halfway through the passage, Tanya reiterated the term insight, and subsequently
described her previous experience of seeming invisible to other people, ironically as a
result of her being quiet. While her response did not provide an answer to the question,
it did convey information about her experience in the world, and her experience of
language, with much of the content of her response semantically primed by the syllable
‘sight’ in the term ‘insight’.
RECOVERY
Recovery is another term that has come to have specific meaning in the mental health
context. This section reports the participants’ own descriptions of the process of
recovery in their lives, frequently occurring spontaneously in their narratives, and also
their responses to a direct question about the meaning of the term in the second
interviews.
A number of participants spontaneously reported strong beliefs about the influence of
the experience of mental illness on their lives. Matthew stated his view outright:
‘I feel I can honestly say that the illness that I went through has made me a
better person, I really believe that Andrew, has made me a better person… I
love my fellow man, I never ever want to do anything wrong, I respect the law,
and I would always like to be a law-abiding citizen’ [Matthew, Interview 2, lines
151-154].
Ethan used the same term, ‘better person’, and expanded on specific aspects that he
identified as being changed by his experience:
Chapter 5 Results
124
‘The other thing is, I think I’ve developed and become a better person because
of mental illness. So I think the mental illness is the main thing that’s prepared
me for the consumer work, you know, it’s sort of, I’m less arrogant, I’m less
narcissistic, less sense of entitlement, I feel more connected, I know what
networking is, and what partnerships are, all of that. Whether I’m good at it I
don’t know, but others seem to think I’m doing a reasonable job’ [Ethan,
Interview 2, lines 211-216].
For Ethan, there was a link between the changes he observed in himself and his
participation in work to help others through the mental health consumer movement.
‘The other thing that really helped was really not to feel superior to other people
that had a mental illness, to actually see yourself as one of them, you know, so
it’s a bit like, “Welcome to the consumer world”, you suddenly realise that
they’re the most helpful people of all, even more than clinicians, although
there’s some very good ones, and also the idea that that I can make choices
you know, I don’t have to be helpless, I don’t have to be at the mercy of other
people’ [Ethan, Interview 1, lines 178-184].
Ethan also described how long the recovery process took, and the elements he
identified in his own recovery:
‘The recovery process actually took up more time than the disintegration, as I
call it, and I really, I’m pretty conscious of the kind of things that helped, being
treated as a person, overcoming difficulties with family, having my family behind
me, my education allowed me to detach myself, so, this is happening to me but
there’s an inner core where I preserve my personality, I think, [and] sense of
humour’ [Ethan, Interview 2, lines 90-95].
The elements he identified include his treatment by others, but also his internal mental
processes, specifically the capacity to detach himself, that drew on his experiences and
attributes prior to developing a mental illness, including his education, his sense of
humour and his sense of an ‘inner core’ to his personality that he could ‘preserve’ from
‘disintegration’.
Michael acknowledged that he had developed personal qualities as a result of having a
mental illness, but he expressed ambivalence about having to experience illness in
order to develop these:
Chapter 5 Results
125
‘I’m optimistic and hopeful that, you know, it’s not going to become chronic
again, and it’s something you don’t want, you don’t want to be the
schizophrenic, even with all, even if it teaches you to have different values and
builds your resilience, there’s no way I’d want to have, no way I’d want to go
through the illness again’ [Michael, Interview 2, lines 207-211].
Michael expressed a positive response to the use of the term ‘recovery’ in a special
mental health context:
‘Oh, as a word I like it, it means that you’re improving yourself, yeah, it’s much
nicer than thinking that you’re a schizophrenic, or have schizophrenia, “recovery
from schizophrenia” sounds much nicer, but in the end it’s no point having
“recovery from schizophrenia” as a label for yourself if it’s not actually
happening’ [Michael, Interview 2, lines 348-351].
His response included specific reference to the capacity for the term to be used as a
label, potentially in a positive sense, as long as it had some basis in reality.
Ellen echoed Michael’s ambivalence about the experience of mental illness:
‘I wouldn’t change anything for the world, but I sure as hell wish it was easier’
[Ellen, Interview 1, lines 754-755].
However she was less positive about the use of the term ‘recovery’:
‘See, that’s one of my, recovery, because then it’s that, that you’re, there’s a
victim, survivor, that you’ve recovered from something, so it’s not one, it’s very
rare for me to use that’ [Shirley, Interview 2, lines 418-420].
Her rejection of the term related to her perceived inference of a person having a prior
history of being damaged, and was consistent with her rejection of identity descriptors
that link to this concept, including ‘victim’ and ‘survivor’. Tanya stated the opposite
opinion, that is, that the broader associations contribute to the benefit of the term’s use:
‘It’s a good word for anyone, if they’ve, like, you know, been through any family
issues or domestic violence or I mean rape or anything like that, recovery’s a
good word for that’ [Tanya, Interview 2, lines 117-119].
Tanya states her own recovery in concrete, functional terms:
‘Oh it feels great, I feel like a new person, I feel good, I feel really, really, really
happy about myself, like before I was really stressed out, you know overweight,
Chapter 5 Results
126
sleeping, no diet, this and that, no money, no smokes, stressing out, scabbing,
bludging [laughing] sorry, but now I put money away, and [I’m] going to go see
my friend in hospital, so yeah, I care for people now, cause back then I was,
you know, people cared for me, and yeah, now I care for other people’ [Tanya,
Interview 1, lines 168-173].
Her description of her recovery included the capacity to help others as a key attribute, a
theme that recurred with several participants.
Shirley, reflecting on how her experience of bipolar disorder affected her life, was blunt
in her opinion of the idea that people would not get rid of mental illness if they could:
‘Stephen Fry’s done his, you know, secret life of the manic-depressive, and he
goes around and asks these people, “If you could push the magic button and
get rid of bipolar, would you do it?” and you know that half, most of them say,
“No, I would keep it”, and I just think they’re idiots, I’ve thought they’re idiots for
a long time, like, what the? I’d push it in a flash, but I think, yeah, I think, kind of,
over a longer period of time there’s more, I guess there’s more acceptance that
that’s kind of been your pathway or your disadvantage or your distressing
events and other meaningful things have happened because of it’ [Shirley,
Interview 1, lines 410-417].
She did note that over time she has come to accept that ‘other meaningful things have
happened because of it’, with the inference that some of those meanings were positive.
For Shirley, meaningful things included her taking on a public role as a mental health
consumer advocate, drawing on her own experiences, and identifying how these
aligned with broader issues:
‘I just think that remembering back to ten years ago, and thinking, a whole
profession hasn’t noticed that depression’s the shitty bit of this thing, you know,
and that just speaks to me that, you know, people with a lived experience aren’t
the central voice at the table and haven’t been, and kind of need to be, no
matter which kind of mental health issue or challenge you’re talking about, so I
guess that’s my sort of particular passion around trying to centralise voices’
[Shirley, Interview 1, lines 669-675].
Shirley’s use of the phrase, ‘the central voice at the table’ is an example of metonymic
chaining. In the first instance, the voice stands for the person, and in the second
instance, the person with lived experience stands for their capacity to represent
Chapter 5 Results
127
opinions about the system. The literal sense of the term also leaves its trace as those
views are frequently expressed through voice. The drive toward embodied
representation is further reinforced by the use of the innocuous word ‘table’, which
creates a specific sense of spatial location.
Ian reported that he had created art specifically addressing recovery from mental
illness, focused on the metaphor of a caterpillar’s transition into a butterfly:
‘I entered an art competition where a caterpillar has five stages of becoming a
butterfly, and I did, you know, the first stages, you feel like a grub, sort of thing,
you’re scraping the bottom of the barrel and you’re walking around all down and
depressed, and then you realise you’ve got to do something about it so you go
into a cocoon, and you develop in the cocoon and come out as a beautiful
butterfly at the end of it, so I am aware of the recovery process’ [Ian, Interview
2, lines 134-139].
His description included further figurative extensions of the primary metaphor; the
caterpillar downgraded via simile to, ‘like a grub’, and the person/caterpillar/grub then
idiomatically ‘scraping the bottom of the barrel’, which uses predicational metonymy to
convey the experience of exhausting all available resources. He went on to describe
the experience of recovery in less colourful terms, highlighting the concept of ‘hope’:
‘Yeah that’s right, like, I believe in hope, without hope you’re really, you’ve got
nothing, I mean you’ve got to have some sort of hope that there’s light at the
end of the tunnel, but it doesn’t come through wishful thinking, it’s got to be a
tangible end to your hope, like it’s got to actually, it’s got to be reality, it can’t
just be a, you know, a wish, sort of thing’ [Ian, Interview 2, lines 157-161].
The term ‘tangible’ was also used by Michael, and functions as a corrective for the
vagaries of inference the other senses endure in the experience of psychosis. Ian also
echoed Michael’s opinion that recovery has to be real and not simply a word.
Ian also talked of the desire to help others as an aspect of his recovery. He articulated
with beautiful economy the experience that lends peer work its power, ‘I was where you
are’ [Ian, Interview 2, line 525], metaphorically and elliptically attributing a spatial
quality to the experience of illness, and metonymically locating himself and another
person in different times in that same space. The effectiveness of this blended
description is not undermined by the fact that, of course, the sentence can also be
Chapter 5 Results
128
interpreted literally, as there can be an actual space, for example, a mental health
inpatient unit, in which the encounter occurs.
Experience of language
This last section reports on aspects of participants’ experience with language that have
not been captured under previous themes. The reported experiences relate to
language within and outside therapeutic contexts and include further data related to
figurative language.
METONYMY: QUICK QUIZ
The responses when participants were asked to determine the inferred meaning of the
question, ‘Would you like another glass?’ demonstrated some of the complexity
involved in pragmatic inferencing and metonymy.
Yves’ response evinced comprehension of the metonymic inference that would be
correct in the majority of contexts:
‘Just being hospitable and saying, you know, “Would you like another glass of
water”, yeah’ [Yves, Interview 2, lines 307-308].
This reading was shared by the majority of participants who interpreted the request
without hesitation as meaning ‘Would you like more of the contents in your glass?’ That
is, they made the pragmatic inference that, in the specific context, the salient concept
the speaker intended to draw attention to was the metonymically inferred contents of
the glass, rather than the glass itself, which is the literal referent of the term. The
capacity to make the routine pragmatic inference was perhaps more notable in that the
question was being asked not in the context of a social encounter where drinking
glasses and their contexts were present, but in an interview which explicitly concerned
language, thus potentially prompting more wide-ranging responses.
A number of participants volunteered additional information about the specificity of the
contents being offered:
‘Well, I’d be thinking Chardonnay’ [laughing] [Shirley Interview 2, line 928]
‘Would you like another glass of beer?’ [Ethan Interview 2, line 382].
The question did not contain additional information about the context, and so in these
responses participants were supplying likely contents, based on their real world
Chapter 5 Results
129
knowledge rather than the immediate linguistic prompt, and linked through chained
metonymies.
Other participants noted the ambiguity in the question, between a literal and a figurative
reading. Tanya observed the possibility of a literal interpretation of the request, but, far
from being troubled by the ambiguity generated, expanded on the attributes of
someone who would think this way:
‘Some people say, “Can I have another glass?” because they’ve already drank
out of it and they’d like a clean [one], they’re poshy people’ [Tanya, Interview 2,
lines 244-245].
Here the desire for clean glassware metonymically stands for the type of person who
might make the literal request.
Matthew struggled to answer the question:
P: ‘Are they offering me a drink, are they offering me a drink?’
I: ‘Hmmm, is that what you think?’
P: ‘I’m trying to think, would you like another glass? I can’t really connect with
that, forgive me, forgive me Andrew.’
[
]
P: ‘As I said, it’s a little bit ambiguous, doubtful in meaning, it’s doubtful in
meaning for me, for me it is doubtful in meaning’ [Matthew, Interview 2, lines
295-299, 310-311].
This was an example of residual difficulty with communication experienced by a person
whose recovery from schizophrenia would be described as comprehensive. Between
interviews he celebrated twelve years since he had been admitted to a psychiatric
inpatient unit, and he does in fact have a social life which includes having the
occasional drink with friends. It is worth noting that, not only did he have difficulty
interpreting a phrase which is regarded as everyday speech, he found the experience
disturbing and socially awkward, as evidenced by his reiteration of the doubtfulness of
its meaning for him, and his repeated request for forgiveness.
While Matthew articulated ambivalence about the question in the context of a ‘quick
quiz’, in his first interview, he spontaneously used the container for contents metonymy,
responding ‘two packets a day’ [Matthew, Interview 1, line 358] when asked about the
Chapter 5 Results
130
extent of his previous smoking. Here ‘packets’ stands not just for the cigarettes
contained in the packets, but an imputed standardized quantity. This in itself is
underspecific, as the time when a packet always contained 20 cigarettes passed long
before he gave up smoking. The point is, he was able to use metonymy in his everyday
speech, but challenged by it when it was posed as a metalinguistic question.
Michael also evinced a sensitivity to the ambiguity between the literal and non-literal
inferences in the ‘quick quiz’ question. This in turn was followed by a tangential
response about communication problems and judgements, then recognition of the
metonymic inference, and then a further comment about communication style,
including clear reference to experiencing difficulty keeping track of context:
P: ‘Oh well yeah, if I take that literally it’s like, do I want another, I’ve got this
one glass, another, it would be another one of those glasses but, [picking up
glass on table]
I: But if we’re sitting here, and I notice your glass is empty and I said, “Do you
want another glass?” what would you think I’d be saying?
P: Yeah, I’m very tolerant, yeah I allow people to talk very, with a low standard,
I’m very, kind of, liberal in that sense, yeah, I try not to judge people if they can’t
communicate properly, because I’ve had communication problems myself,
sometimes I’ll be like halfway through a conv-, a sentence and the other person
will have to jump in and I can’t even fi-, end the sentence. I try to like, think
things out too much, [laughs]
I: Is that generally or is that?
P: Yeah that’s generally, so if you said, “Do you want another glass?” I’d know
what you meant, you mean do I want another drink yeah,
I: Without thinking about it?
P: Well that’s the thing, I operate from a knowledge centre [pointing to chest]
more than a thinking and analysing thing, yes, so often I’m caught out because
in this short period something’s changed, and I haven’t really kept up with the
change’ [Michael, Interview 2, lines 122-140].
Like the inferences about the specific content in the proffered glass above, these
comments about communication difficulties were not directly prompted by the question,
but spontaneously expanded by Michael. Michael also clearly articulated the
Chapter 5 Results
131
experience of missing a contextual cue, and subsequently not being able to keep up
with a conversation.
Ethan reported detecting a threat in the question:
‘I could think of you saying that, and I could think “glassing” but just, sort of, first
word in the head it triggered’ [Ethan, Interview 2, lines 384-386].
This use invokes the metonymic transfer of the object glass into the assaultive act of
striking someone with a glass; this is an atypical inference, again, drawing on world
knowledge not directly prompted within the provided language, and suggestive of
hypervigilance.
In summary, many participants demonstrated comprehension of the intended inference
of a referential metonymy with minimal context provided. Several participants
formulated additional inferences, both positive and negative, based on extra-linguistic
knowledge. Several participants noted the presence of ambiguity, with the presence of
the source meaning not necessarily yielding entirely to the target meaning. For two
participants, this ambiguity was experienced as problematic.
METONYMY, METAPHOR AND OTHER LANGUAGE PHENOMENA
This section contains salient examples of participants’ use of both routine and novel
figurative language. These have been selected, not because they illustrate the themes
identified above, but because they furnish examples of the use of language observed
by earlier researchers in mental health and linguistics. They demonstrate both skilled
and idiosyncratic language competencies.
Tanya spoke of her emotional response to feathers, illustrating the capacity for
metonymies to be linked together in chains:
‘I don’t like feathers because I don’t like, you know, people taking them off the
bird and that, I think it’s sad, yeah, reminds me of death’ [Tanya Interview 1,
lines 210-212].
The feather stands not just for the bird from which it was plucked, but for the bird’s
death, perceived as a necessary prelude to the plucking. In doing so, metonymically,
she revealed a darker aspect behind what is normally imputed solely as beauty.
Tanya’s speech demonstrated a number of distinct language patterns:
‘I guess when I get my own apartment, I’ll take the keys with me in case
something happens to Mum or I need to do my laundry, cause I’ve got the keys
Chapter 5 Results
132
for the house, I got it cut down at the key-cutting hut underneath, on the ground
floor, underneath the escalators, got them cut and it’s got some sort of little fairy
on it, but I keep them so that I can look after her still, like drop in time by time
and if she needs food or things done, just to check on her that she’s alright, to
see if she’s still living, breathing’ [Tanya, Interview 2, lines 47-53].
The passage begins as a discussion of her reasons for keeping keys to her mother’s
apartment, followed by details of where she got the key cut, and the specifics of the key
ring, before returning to the reasons for holding the keys. This is a classic example of
circumstantiality, where the speaker starts to include details that are extraneous to the
topic, then goes further into those details, before returning to the original topic without
external prompting. At the same time as close details of the spatial location of the keycutting hut within its setting, the overall setting is not specified, which would be a more
typical attribute for selection as salient. The passage ends with an example of
predicational metonymy, where ‘breathing’ stands for ‘living’. In this instance, the
metonymic term is redundant, as she has already used the superordinate term. This
demonstrates that metonymy is not always used solely to make communication more
economic, but also to create inferences and nuances in meaning.
Ian used predicational metonymy in a familiar way to many of us when we can’t
remember an actual name, adopting salient attributes to establish the identity of a
public figure:
‘That girl, who jumped off the gap, who was on Channel Ten news’ [Ian
Interview 1, lines 530-531].
In fact he ascribed two properties, as each on its own would be underspecific to identify
the individual person, but both together are sufficiently specific to a listener who had
been exposed to media over the relevant period. Again though, seemingly
straightforward descriptions use metonymy and ellipsis to convey the full message.
‘Jumped off the Gap’ elliptically described her suicide, as ‘the Gap’ is a common name
for a specific cliff on Sydney’s coastline, and jumping off it almost invariably results in
death. Further, ‘on Channel Ten news’ here means she was a newsreader on Channel
Ten, though there is nothing in the actual words used to distinguish this from any
person who may have been featured in a story on Channel Ten news.
Participants frequently used predicational metonymy, with no notable difference from
the way it is typically used. For example, Ethan stated, ‘I managed to stay out of
hospital pretty much’ [Ethan, Interview 1, lines 76-77] where his ability to stay out of
Chapter 5 Results
133
hospital stands for his actually doing so. Ethan also stated, ‘you might need to jog my
memory’ [Ethan, Interview 2, line 9], and again, the insertion of ‘might’ almost passes
unnoticed. The fact is, he did need to be reminded, but he refers to this as a potential
rather than a real state. Other participants also used this type of metonymy, for
example, Tanya, when asked how she describes her experience, stated, ‘I’d have to
say I’ve got schizophrenia’ [Tanya, Interview 1, line 121] with obligation to say standing
for actual saying. Similarly, Nathan stated, ‘I was able to be released’, and Matthew
used ‘I must admit’ [Matthew, Interview 2, line 118]. Participants demonstrated use of
predicational metonymy in their speech production without any of the opacity or
atypical use that was occasionally noted in their use of referential metonymy.
Ellen’s speech also featured a range of uses of distinctive linguistic practices, including
reported speech, repetition, and mixing of elements from different discourses:
‘When XX, my now ex-husband, called and said, “Ellen has been hit by a car
and she can’t come in to work today.” my manager said, you know, “You’re
joking, right?”, she just didn’t believe him because we had discussed, like that
morning I came in, we talked wedding and Christmas, and then we focused on
vision and strategy and you’re all on the same page and she said, “I took a
deep breath.” and I thought, I can’t believe it, this is everything I’ve worked
towards is here, here and now, yay, three minutes from home, you know,
stepped off a bus, crossed the street, I remember the green walk sign, and that
was it, my life changed’ [Ellen, Interview 1, lines 354-361].
In the first instance, she used direct speech to describe a telephone exchange between
her then husband and her manager, even though she would not have been able to
hear, let alone recall the exact words 12 years later. She then takes a step back
chronologically, and narrates the morning prior to a car accident, and her description
used terms drawn from corporate discourse, ‘vision and strategy’ and ‘on the same
page’, the latter an idiom in which people observing the same page in a document
(which may be real or metaphoric) metonymically stands for people being in
agreement. In the second-last line of the passage, Ellen used a predicational
metonymic description for the location of an accident, ‘three minutes from home’. It is
clear that she was describing the place, that is, the location in space where it occurred,
as the other descriptors are consistently visuo-spatial (e.g. ‘green walk sign’). But she
did not state, ‘200 metres from home’ which would be the literal means of conveying
location information, but instead metonymically substituted the estimated time to walk
that distance to describe it. This is an example of a type of metonymy that is so deeply
Chapter 5 Results
134
ingrained within the everyday way of conceptualising time and space that it generally
passes unnoticed.
Ellen described being placed on a community treatment order as ‘I had to go through
parole’ [Ellen, Interview 1, line 233], ironically equating the legal obligations placed on
her for some duration to the legal force placed on people who have been convicted of a
criminal offence and subsequently released from prison. This is a clear instance where
the trace from the source is intentionally foregrounded in naming the target. While
offered ironically, this term clearly indicates that being subject to mental health
legislation is experienced as equivalent to being punished as a criminal.
Michael used another frequently encountered concept, the metaphor of an illness being
a ‘journey’, several times and when asked to speak further about it, stated,
‘Well it’s just change, I mean, even looking back before I became unwell, it was
a change becoming unwell, so it was a journey even though it was an unwanted
journey, travel through that’ [Michael, Interview 2, lines 224-226].
This demonstrated that, while his initial use of the metaphor may not have been overt,
he was readily able to expand on potentially available inferences in its use. For
instance, he noted that ‘it was an unwanted journey’. This is an example of use of
figurative language where there is a disruption between the source and the target, as
the presumptive inference is that a ‘journey’ is undertaken voluntarily, thus requiring the
addition of a qualifying adjective. He then metonymically extended the journey
metaphor by saying, ‘travel through that’, which incorporated the idiomatic sense of
overcoming adversity as getting through something. Soon after, he offered a further
observation on the ways that people can conceptualise the experience of illness:
‘Well I don’t know like, archetypes and Joseph Campbell type stuff, everyone
has their own story, they’re the hero in their own story, maybe that’s not cool to
call it that, that’s how I perceive it’ [laughing] [Michael, Interview 2, lines 230232].
Michael linked his use of the term ‘journey’ to the writings of the mythologist Joseph
Campbell, suggesting a pattern of deeper reflection in his choice of the term.
Michael used the only example of illocutionary metonymy, when he spoke about his
relationship with his mother, he opened with ‘[I] should probably mention my mother as
well’ [Michael, Interview 1, line 102], where the presumed obligation to speak about his
mother stands for him speaking about her.
Chapter 5 Results
135
NARRATIVE
Examples of participants’ use of different narrative modes have been noted within
previous sections. This section reports on participants’ varied, and occasionally
contradictory, ideas about narrative.
Three of the interview participants have worked in roles where they have previously
narrated their experience of mental health issues and subsequent treatment in various
instructional contexts. These include presentations to other people experiencing mental
health issues, to practising clinicians and students in undergraduate courses, and to
audiences of first responders, for example, police and welfare workers.
For other participants, narrating their stories was a novel experience. At the end of the
first interview, I asked Tanya, ‘Are there any other stories that you want to tell?’ Her
response subverted my inference that the story be about her experience of mental
illness or recovery:
‘Well I, I found these, I found a dead mynah bird in my front yard, and five years
ago I found another dead mynah bird in the lion head pond, I go there because
the lion is my star sign, and I used to put flowers in there. Anyway, they’re five
years apart and they’re both buried in the backyard. I went up to the lion head
pond and I put a flower in there the other day, then like, the next week later I
went back to put another flower in there where I found the first mynah bird, and
the other flower was still there, fresh, because it was in water, yeah, and I was,
when I first buried them, they’re five years apart, I put flowers on their grave and
stones around it, and covered it with stones, yeah’ [Tanya, Interview 1, lines
192-200].
Instead, Tanya offered a brief narrative, outlining her relationship to nature and space.
The ‘lion head pond’ is a specific place, with a stone replica of a lion’s head as the
outlet for a fountain. Speaking of her motive to go there as linked to her star sign
indicated an intensely personal relationship to meaning, the linking of the salient
elements functioning as motivation for action, not just description. The passage also
demonstrated her relationship to other life forms, with near magical thinking, as in the
flower staying alive in the water for a week, and the linking of the two dead birds over
five years. Her respect, in arranging tributes to the dead birds, was also evident. The
whole passage spoke of a life where her experience is invested with meaning.
For Ethan, narrative itself functioned to create meaning from experience. He linked the
concept of deliberately crafting a narrative to create meaning, metaphorically conveyed
Chapter 5 Results
136
as ‘weaving’, with having insight, thus making insight a dynamic process rather than a
fixed state:
‘It’s also very useful to reconnect the chaotic last forty years into some kind of
narrative or story, and if you don’t have insight, I mean, in that sense of weaving
together your life into some kind of meaningful thing instead of a slag heap, it’s
very difficult’ [Ethan, Interview 2, lines 259-262].
Ethan is one of the participants who has undertaken consumer advocacy work that
includes the use of his own narrative with different audiences. He described the
process as always evolving, in that, in reconstructing his story for new purposes, he
discovers additional meaning which he incorporates into his own process of recovery.
Shirley has also retold her story publicly, but her reflections on the process were
different from Ethan’s, extending to a reflective critique of this use of narrative:
‘There’s some people who are really trying to talk about this idea out of Canada,
called, “patient porn”, which I think is just fantastic and I use it to critique my
own kind of storytelling: “At what point is the telling of these experiences just
simply kind of spectacle?” and that’s something I’ve been thinking of a lot
actually at the moment’ [Shirley, Interview 2, Lines 859-863].
Her account shared with Ethan’s a sense of narrating past experience as a dynamic
and reflective process. However, her question rested more on the role that these
narratives play for the audience rather than for her, with details of her life offered up for
voyeuristic consumption. Shirley linked this in part to a frequently observed
phenomenon whereby narratives serve explanatory functions:
‘People have this innate need to make sense of it [mental illness] and attribute it
to something, which I’m actually not that interested in to any, I’ve got about ten
percent interest in that, relative to other people’s 100% interest in that, so that
sense-making process is important to other people, and the attribution of cause
seems to be very important in, to people at large I’ve found, so in my work I
definitely use things like, “Well, there were a range of things going on”’ [Shirley,
Interview 2, Lines 749-754].
As part of her reflections on her story and the purposes to which she can put it, and to
disrupt what she saw as other people’s narrow interests, Shirley adopted a dual
strategy, suggesting a multiplicity of factors that influence developments, rather than a
Chapter 5 Results
137
single causal element, and then underspecifying those factors within her narrative of
her own past, using the deliberately vague term, ‘a range of things.’
Ellen articulated a tension inherent within a narrative interview in relation to what is
sought: an answer to a question, or a narrative? ‘I also forget what the hell you ask me
because I go into my story’ [Ellen Interview 1, lines 334-335]. Her response reflected
the idea that her story exists on its own terms, and she tells it in that way, rather than it
being constrained to be ‘fit-for-purpose’.
PARTICIPANTS’ OWN OBSERVATIONS OF THEIR USE OF
LANGUAGE
Ellen described having a longstanding interest in language, and commitment to using
language deliberately in order to communicate well. She reported a response she once
received from a vocational worker to her impressive vocabulary: ‘Did you swallow a
dictionary when you were a kid?’ [Ellen, Interview 1, lines 744-745] and she described
always making conscious choices in her language production:
‘When I engage with someone initially I ascertain, like, use different words, and
I’ll say, “Ooh great, oh great I can use my vocab,” or, you know, just simplify it,
and I’ll always check in with someone ‘cause I don’t want to, I’m not interested
in dumbing down language, but it’s about what someone’s comfortable with’
[Ellen, Interview 2, lines 145-149].
She talked about the importance of using skilled communication in her interactions with
mental health workers: ‘I need to be able to articulate in a way that I am taken seriously
and not seen as a number’ [Ellen, Interview 2, lines 88-89]. In doing so, Ellen
challenged the practice of referring to health care recipients as numbers. Ellen also
reported that, alongside her practice of checking that other people can understand her
speech production, she is also comfortable seeking clarity when listening to someone,
ensuring her speech comprehension:
‘I don’t make assumptions, if it doesn’t, if it doesn’t flow on with what’s being
discussed, if it’s just something different, I’ll check in, I don’t have any issues
with clarification’ [Ellen, Interview 2, lines 249-251].
Ellen reported her evaluation of the result of this care in communication:
‘I know that people trust me because of my use of language, and everything
that goes with that’ [Ellen, Interview 2, lines 238-239].
Chapter 5 Results
138
In Ellen’s assertion that people trust her, her ‘use of language’ is both the salient
element in itself, but also stands for ‘everything that goes with that’, its power thus
operating both literally and metonymically.
Other participants reported awareness of difficulties they experience with language.
Michael has achieved significant control of the positive symptoms of hallucinations and
delusions. He is motivated, has studied at university, works and has spoken publicly, at
times in front of audiences of 200 people. Nonetheless, he experiences communication
difficulties:
‘It happens to me a lot of the time, like, I just don’t have anything to say with
people that I’m with, and then when I do say, it’s in some sort of vague general
way that, I just give up sometimes talking. Yeah, so that, I think I haven’t really
honed my communication skills because of the psychoses, and the illness for
like, a decade, so I’m just trying to pick up the pieces and improve’ [Michael,
Interview 2, lines 329-333].
For Michael, his experience of problems with speech production is related to his
experience of psychosis over more than a decade, with the implication that it is
specifically a problem of social skill. His experience echoed Tanya’s report that she
‘just stopped talking for five years’. Michael’s response to an earlier query whether he
ever had the experience of not comprehending another person’s sentences or phrases
revealed a rigid concept of language, ‘Well, that’s what the dictionary’s for’ [Michael,
Interview 2, line 120], suggesting that his self-reported problems with language may not
be restricted to lost opportunities in the social realm, but may also include a language
processing issue that impacts directly on his capacity to form pragmatic inferences.
This is reinforced by his earlier report of not keeping up when meanings change in
specific contexts.
Yves recounted an episode of communication difficulty he experienced as a mental
health inpatient:
‘Talking to other people in hospital sometimes we’re all off track, so we’re all off
track with each other, so we’re having, heaven knows what we’re having
delusions of, and we’re all sort of talking off the air’ [Yves, Interview 2, lines
243-245].
His report was of people concurrently experiencing derailment. Asked how they
responded to this situation, he explained:
Chapter 5 Results
139
‘Oh, you sort of like, go and have a smoke together or something but you just
don’t really notice it, you notice it but you just ignore it, you just sort of, it’s just
part of the, sort of, lifestyle’ [Yves, Interview 2, lines 337-339].
Yves reported a generally good-humoured response to this, ‘hav[ing] a smoke together’
metonymically suggesting the breakdown in communication did not necessarily lead to
a breakdown in friendly relations. When asked if he thought there was a benefit in
clinical staff drawing attention to derailment when it occurred, Yves firmly expressed
the view that it would not be useful.
Shirley’s ideas on language use in recovery also included an important distinction
between language used with peers and language used with mental health workers:
‘One of the most critical things in that was language actually, because about 18
months prior to that I’d spoken to someone who’d had some similar
experiences, and they referred to those types of experiences as ‘scary
thoughts’, and I just kind of, I started to use language, just internally, that just
tried to stop using the catastrophe language and stop seeing it as a crisis and
stop seeing it as psychosis and I tried to bring it down to, sort of, matchbox size
and go, “You’ve just had some scary thoughts”’ [Shirley, Interview 2, lines 900906].
Shirley accorded language a significant role that recalled the function of narrative for
other participants, speaking of ‘sense-making’ where others had used the term
‘meaning’. Shirley expressed the complex idea that language cannot adequately
capture reality, yet it may contribute to shaping reality:
‘I think that some of the language we use just isn’t powerful enough for some of
the experiences we have, and then I think, just that idea that people can be
supported and encouraged to explore language use for themselves as part of
healing, you know, as part of sense-making of what’s happened, and you know,
I did have what I think is the equivalent level of intense experience last year as I
did fifteen years ago, but one of the big differences around bouncing back from
that has been my ability to shrink it down using language, and my experience
making sense of it through conversations with peers, not with health
professionals, and just my own reflection on that’ [Shirley, Interview 2 lines 9941001].
Chapter 5 Results
140
In particular, Shirley spoke of being able to ‘shrink down’ the intensity of an experience
through using language. By using the term ‘ability’, Shirley highlighted that her use of
language was something she could control, and that this could in turn contribute to her
capacity to control her responses to her psychotic experiences.
Shirley, through her strategic use of language, further demonstrated a purpose that
went beyond her own situation, when she articulated an effective argument about the
responsibility that psychiatrists have to monitor the physical health of the people they
treat:
‘If your pen’s going on that prescriber’s pad, you are the one that needs to read
the journals about the physical health impacts, and you should know what’s
going on with my kidneys, or my whatever, my metabolic rates and stuff like
that’ [Shirley, Interview 1, lines 712-715].
The ‘pen [ ] on the prescriber’s pad’ struck me as a powerful use of metonymy as soon
as I heard it in the first interview. The responsibility inherent in the practice of treating
mental illness, from diagnosing to treating, was metonymically reduced to one salient
aspect, applying a pen to a prescriber’s pad. When, in the second interview, I asked
Shirley to comment on her use of this term, she reported that, in her role as a
consumer advocate, she had been asked several times to provide comment on mental
health issues to the press. Shirley reported she had been told by one journalist, ‘I can
give you two sentences, maybe three.’ [Shirley, Interview 2, lines 833-834], so she
learned to make her communication more economic. In this context, her use of the
phrase proved successful. This relates to the original role of metonymy in rhetoric, to
persuade.
Conclusion
The information provided by participants demonstrates complexity both in their
experience of mental illness, and their experience of language. At times, complexity
arose from the intersection of these experiences.
METONYMIC INFLUENCES ON PARTICIPANTS’ EXPERIENCE
The influence of metonymy was demonstrated across several aspects of participants’
experience. Participants discussed their thoughts about identity; all reported they had
thought about the topic, and for all participants, the experience of being diagnosed with
a mental illness influenced their sense of identity. For some, there were benefits to this,
and several participants spoke about how they felt their experience of mental illness
Chapter 5 Results
141
had made them better people. Participants also reported that having a diagnosis gave
them a sense of control. For others, there were negative aspects to adjusting to living
with a mental illness, and the necessary realignment to their previous sense of self.
These aspects had material consequences in people’s lives, influencing their
interactions with others, and their capacity or willingness to accept treatment.
One strategy participants reported to address this risk was reclaiming other attributes
as significant in their identity formation. These included resuming roles like caring for
family members, working, studying and undertaking peer advocacy work. These
strategies were related to a key concept underpinning identity, visibility. While
participants reported their identity formation involved self-reflection, it was also strongly
influenced by the image of themselves that they saw reflected by others. In this
context, it follows that undertaking socially recognised activities like working and
helping others supports a positive sense of identity. Other participants reported that it
was the invisibility of their symptoms that led to people not understanding, or otherwise
underestimating their experience of illness.
All participants reported experiencing stigma. This was frequently, but not exclusively
related to the perception that they had a mental illness. Both participants who were first
diagnosed with mental illness in the 1970s expressed the view that stigma is getting
worse.
Participants reported negative experiences, including poor communications, with
mental health workers. In particular, they reported the experience of not being listened
to, and having their explanations of events, and their wishes and values discounted.
Participants reported being denied services as a result of their mental illness, and
conversely, as a result of their mental illness not being recognised. This included
having physical problems undertreated or dismissed when their mental health history
was noted. These forms of discrimination are based on stigmatising attitudes about
what people who live with mental illness are like, and the credibility of what they say.
Participants also reported positive aspects of the mental health treatment they had
received. Positive elements included workers who fully engaged with them as whole
people, without foregrounding their mental illness. This worked in reverse as well, with
participants noting the best interactions were with health workers who acted as whole
people themselves, not restricted to their clinical roles. This engagement was
frequently evident in the language that was used in clinical encounters, as reported by
participants.
Chapter 5 Results
142
METONYMIC ASPECTS OF PARTICIPANTS’ LANGUAGE
In terms of speech comprehension, participants demonstrated the capacity to interpret
metonymic speech used by others, including the metalinguistic capacity to play with the
language. Participants demonstrated their enjoyment of this capacity to juggle multiple
potential intended meanings. Several participants demonstrated difficulties with
comprehending figurative language. In relation to speech production, several
participants also demonstrated idiosyncratic speech patterns, including elements
suggesting the presence of thought disorder. The presence of these difficulties created
material effects on people’s lives, with some participants reporting they ceased or
reduced communication with others because of these problems.
When describing their symptoms of mental illness, participants frequently used
figurative language to convey their experiences. They also used narrative skills in
telling their stories.
Participants demonstrated critical awareness of the language used about mental health
in general, and themselves in particular, in the language they encounter from mental
health workers and the broader community. They have also demonstrated strategies to
manage the influence of this language. The experience of language as reported by
people who live with mental illness will be explored in relation to previous literature in
the following Discussion chapter.
Chapter 5 Results
143
Chapter 5 Results
144
CHAPTER 6: DISCUSSION
Two research questions were introduced at the beginning of this study:
How does the use of metonymic language influence the experience of people who
live with mental illness?
How does metonymy influence the speech production and speech comprehension
of people who live with mental illness?
The ubiquity of metonymy in everyday speech (Lakoff & Johnson 1980; Littlemore
2015) was evident in the language experiences reported by participants. Metonymic
language is used by them, and about them. In particular, the influence of labelling,
previously identified by social theorists (Goffman 1963, 1968; Scheff 1974) is still
prevalent. Labelling enacts metonymy, and although these concepts have not
previously been linked in the literature, the data from the interviews demonstrated their
co-occurrence.
The evidence from participants’ narratives will be considered in light of previous
literature. Mapping the influence of metonymy on participants’ reported experiences,
the chapter begins with a discussion of identity, which was an a priori theme for the
study. The way in which identity is metonymically constructed through the selection of
specific attributes is addressed. This is followed by a discussion of participants’
description of the visibility or invisibility of mental illness, underlining that identity is
constructed within a social context. In cognitive linguistic terms, the visibility/invisibility
distinction rests on the process through which an aspect of a phenomenon is
metonymically selected for foregrounding. This concept is considered alongside
literature about marginalised groups for which membership status is not necessarily
obvious. The following section addresses the experience of stigma, which was reported
by all participants. This includes reported experiences of poor treatment when
accessing health services, which participants perceived to be related to their mental
illness. In this context they also reported some uncertainty about the inferences they
drew from their experiences. This uncertainty is related to the way in which metonymy
can mobilise two meanings at once, a literal and a figurative one, each of which carry
different potential inferences. The chapter then addresses participants’ reports of their
experience of symptoms, which echo previous suggestions in the literature about
possible metonymic motivation for delusions.
Chapter 6 Discussion
145
The words that participants use to describe their symptoms demonstrate figurative
language, and this leads into a section on metonymic influences in the speech
production and comprehension of participants. The evidence for the concept of
metonymic speech in psychosis, first put forward in the 1930s, is considered. An
exploration of participants’ speech comprehension follows, followed by a discussion of
the range of uses of metonymy in participants’ speech that are consistent with
everyday language practice in the broader speech community. The metonymic
strategies that underpin the function and operation of narrative in participants’
interviews are identified.
Finally, the chapter addresses the study’s contribution to cognitive linguistics, the
limitations of the current study, and its implications for research and practice. The
conclusion draws together the concepts explored within each research question.
The influence of metonymic language on the experience of
people who live with mental illness
Participants reported that the way that language is used and interpreted in the varied
contexts of everyday life and in clinical encounters has a marked effect on their
experiences of living with a mental illness. The broad answer to the first research
question is that the influence of metonymic language is experienced in both positive
and negative ways.
Participants made inferences about the world as they experience it. They reported that,
in turn, inferences were made about their language, their appearance and their history,
by clinicians and others. These inferences had material effects in their lives, including
the experience of services being delivered coercively, or being withheld. Metonymy
underpins this pragmatic inferencing; in some cases participants selected which
attributes or elements of a situation would be foregrounded, and which would be
relegated to the background. In other instances, they reported these processes being
enacted by others.
IDENTITY
The onset of mental illness typically occurs in early adulthood, and the novel
experiences can radically alter people’s sense of self, and expectations for the future
(National Mental Health Commission 2014). This is also a period when people are
questioning their identities, often contesting premises about identity based on
membership of social categories. Gender, sexuality, class, political and religious beliefs
Chapter 6 Discussion
146
are some of the attributes that can be foregrounded in people’s sense of self, as can
the experience of illness. Each of these categories can be embraced or resisted (Butler
1990).
All participants reported that identity was an important concept for them and that
incorporating the experience of mental illness into their identities had proven a
challenge. They reported having little or no knowledge about mental illness when they
were first diagnosed, and frequently their images of what it meant to experience mental
illness were overwhelmingly negative and stigmatising; as Shirley put it, ‘you’ve just
been given the diagnosis of doom’. This lack of knowledge, and subsequent fear, was
typically represented as shared by their families and social networks. At the same time
as they were attempting to incorporate knowledge about mental illness into their
identities, they were also relinquishing their previous sense of identity, built on
attributes such as intelligence and competence, and marked by badges of success
including jobs and relationships. Viewed cognitively, the onset of mental illness
metonymically displaces other attributes, whether they will it or not.
Participants spoke of identity in different ways. For some, identity was significantly
constituted through specific attributes, while for others it existed independent of such
ties. The functional role of identity as a factor explaining the motivation for social action
was reported. A powerful identity narrative articulated spontaneously by several
participants was that, although they did not enjoy all aspects of experiencing mental
illness, they believed that the experience had made them better people. They reported
that their sense of identity became stronger and more compassionate because of their
experience of dealing with the adversity of mental illness. They spoke of these changes
in abstract terms including values, beliefs, and resilience, but they also reported
developments in pragmatic terms, including the renewal of their capacity to care for
others. This practice enacted values that were important to them internally, but also
involved performing a role that was visible to, and validated by others.
Participants described a path, from reluctantly accommodating a stigmatised element
in their identities, towards accepting the value the experience has added to their lives.
The participants in Barham and Hayward’s (1991) study of people living with mental
illness in the first decades post de-institutionalisation voiced similar changes in their
values, including a shift from desire for money and material things to a greater
appreciation for self and recognition of others. This shift is paralleled by reports from
other groups who have received diagnoses that carry stigma:
Chapter 6 Discussion
147
HIV-positive interviewees described the identities that took them over after
diagnosis as fabricated out of stigma, but progressing towards acceptance,
even affirmation, of their condition, and towards an enhanced awareness of
what is important in life (Squire 2000, p. 200).
The process of being negatively identified by means of their mental illness was
subsequently overridden, as the manner in which they have met the adversity became
an attribute available for selection in the formation of positive identity. A further
successful strategy reported by participants was to incorporate the experience of
mental illness into one’s identity, and highlight the positive value inherent in
overcoming adversity. Michael expressed this as the capacity for everyone to be the
hero of their own journey. These comments resonate with those of the former National
Mental Health Commissioner, Janet Meagher, (2014, p. 8) who recently stated:
Anyone who can find their way through the personal mess and the emotionally
charged losses that accompany a life lived with schizophrenia, is, in my
judgement, a real hero. The singular effort and courage involved in living and
surviving from day to day is, without exception, a hero’s journey.
Participants’ descriptions of their identity also voiced cultural understandings of the
concept, in particular structural ideas that posit a ‘core’ at the centre of identity. In
tracking the historical development of these ideas, White (2007, p. 102) has identified
that they include:
The evolution of the concept of a "self" is an essence that is understood to
occupy the centre of a personal identity. Although this idea of a self is a
relatively novel idea in the history of the world's cultures, it has been a hugely
successful idea and is today quite taken for granted in the West.
As Flaskas (2002) has noted, however, while critique of such structural notions of
identity may support therapeutic approaches, for many people, maintaining a coherent
sense of identity is a more important goal than challenging philosophical ideas about
selfhood.
The narratives demonstrated that participants had given thought to the idea of identity,
and adopted a number of strategies for incorporating the experience of living with
mental illness into their own senses of self. They also demonstrated that identity
formation is not simply a process of self-selecting which aspects they consider to be
significant about themselves, and foregrounding these. Rather, identity is a concept
Chapter 6 Discussion
148
that exists within a social context, mediated by contending forces beyond deliberate
control of any individual or group. Within this context, the metonymic process by which
salient attributes are identified for attention, with consequent influence on the
inferences drawn, was described by participants in terms of visibility.
VISIBILITY
For most of the participants most of the time, the fact that they had experienced mental
illness was not discernible to others. This unmarked status carried both benefits and
problems. Shirley and Ian both noted that the extent of the suffering people
experienced from mental illness was invisible to the general community, and that the
courage it takes to deal with mental illness was rarely afforded the kind of recognition
that people coping with physical illness receive.
In this context, participants frequently contended with decisions about whether to
reveal their experience of mental illness or not, to ‘put a pole up and raise the flag,’ in
Tanya’s words. Michael, Ethan and Shirley have each made the decision to undertake
roles as public mental health peer advocates. Having once chosen to reveal their
experience in the public domain, they no longer control who can or cannot know.
In terms of both identity and visibility, these experiences align with the ‘Person (mental
patient) / Mental Patient (person) predicaments’ previously identified by Barham and
Hayward (1991, p. 144). Negotiating these predicaments entailed ongoing decisions for
participants about which aspect of the self should be chosen for highlighting in any
specific context. In cognitive linguistic terms, a speaker can choose what is the focus of
attention by metonymically foregrounding a specific attribute of an entity (Talmy 2007).
The decision to reveal, or not reveal, an aspect of oneself enacts Langacker’s (1993)
concept of reference-point phenomena. Selecting what the reference point will be
determines the inferences that are drawn. The operation of cognitive processes in the
focussing of attention was described more than a century ago:
But we do far more than emphasize things, and unite some, and keep others
apart. We actually ignore most of the things before us (James 1890/1950, p.
284, underline in original).
What people choose to attend to and what they choose to ignore each contribute to an
understanding of the issues identified by participants. A further aspect of the
predicaments identified by Barham and Hayward (1991) is that, notwithstanding the
selection of one attribute for attention, elements that may subsequently be shifted to
the background nonetheless leave a trace, intended or not. This aligns with the
Chapter 6 Discussion
149
potential for metonymy to be ‘an efficient way of saying two things for the price of one’.
(Brdar-Szabo & Brdar 2011, p. 236). Barham and Hayward (1991) stress that it is
important to maintain awareness of the difference experienced by people who live with
mental illness, and not elide recognition of this, even as their personhood is
foregrounded. Metonymy contributes to the cognitive capacity to maintain this dual
awareness.
Participants reported many situations in which it was not clear if their experience of
mental illness were known or not known to others. Moreover, this opacity extended to
whether the knowledge would have a functional outcome. Participants in Barham and
Hayward’s (1991) study similarly spoke of the sense that other people probably knew
they had mental health problems, and that this was not an issue when, for example,
they were meeting for a drink, but would become an issue, albeit not openly
acknowledged, if they were to apply for paid employment. In this light, participants’
reported experiences of work are illustrative. Nathan enjoyed the mark of a specific role
‘as a roadie’ in which his experience of mental illness was not a factor. Shirley, on the
other hand, masked her mental illness behind office banter in order to retain her job, a
mask that was reportedly so successful with colleagues it marked her as the type of
person who it is presumed cannot have experience of mental illness. She described
this experience as ‘passing’, and expressed clear recognition of the benefits and risks
involved. Her capacity to present as chatty even when depressed functioned as a
protective factor; however, it also exposed her to hearing stigmatising attitudes
expressed by fellow employees toward others with mental illness.
Passing has a long history with members of other marginalised groups who may not
immediately appear to belong to a marked minority. At the turn of the twentieth century,
the term was used by people of African descent who were light-skinned, and would
selectively ‘pass’ as white for various social ends. This phenomenon was explored in
the novel Passing (Larsen 1929). Butler (1993, p. 170) makes explicit the complexity of
how passing functions in that specific narrative, and more generally:
Blackness is not primarily a mark in Larsen’s story, not only because Irene and
Clare are both light-skinned, but because what can be seen, what qualifies as a
visible marking, is a matter of being able to be read a marked body in relation to
unmarked bodies, where unmarked bodies constitute the currency of normative
whiteness.
Chapter 6 Discussion
150
Thus, an attribute such as skin colour, metonymically connoting race, may or may not
be noted, depending on its importance in the context of marking it as other to the norm.
Whether an attribute is attended to or not depends on factors other than the degree of
difference it embodies, and this attention may be influenced by social or political
categories that may be mobilised strategically to further particular interests.
Passing has historically allowed people to avoid the explicit persecution directed at
other members of their communities. This bears out Goffman’s (1963, p. 74) comment:
Because of the great rewards in being considered normal, almost all persons
who are in a position to pass will do so on some occasion by intent.
Passing has also been practised by members of the Lesbian, Gay, Transgender,
Bisexual and Intersex (LGTBI) communities, whose status may be marked or
unmarked, but has traditionally carried the risk of stigma and violence if known. A
strategy adopted within these groups was to present in public accompanied by a
member of the opposite sex, thus creating a presumption of heterosexuality. These
companions were known as ‘beards’ by insiders. Hardie (1999) has analysed how the
term beard originates as a metaphor for natural masculinity, and is then metonymically
deployed, as ‘standing for’ heterosexuality, and available to both gay men and
lesbians.
Ethan reported adopting an alternate series of what Goffman (1963) termed ‘spoiled
identities’, including drunk, dropout and hippy, as all were preferable to being ‘seen as
a lunatic’. Ethan’s assessment was that the strategy failed, and that ‘people thought I
was crazy as well as drunk’. Ethan’s strategy directly enacts a process formerly
identified by Goffman (1963, p. 94):
Another strategy of those who pass is to present the signs of their stigmatized
failing as signs of another attribute, one that is less significantly a stigma.
In a similar way to the ‘beards’ formerly used by LGBTI people, the use of these
substitute identities operated in a dual way. The strategies Ethan mobilised to disguise
the devalued attribute of mental illness actually drew attention to their deployment as
disguise. This reinforces the idea that the selection of a particular attribute for attention
is not a straightforward matter, and is often determined by other social forces operating
within the context.
Chapter 6 Discussion
151
STIGMA
All participants reported experiencing stigma. This included having services withheld or
delayed, services delivered coercively, and services delivered without clear
explanations being shared with participants. Participants also reported self-stigma.
Stigmatising language frequently made explicit the link to participants’ perceived
mental illness. Two of the terms cited by participants that were used against them with
hostile intent, ‘loony’ and ‘psycho’, are shortenings of terms that have been, or still are,
used clinically, ‘lunatic’ and ‘psychotic’. These terms have undergone a transfer in
inferred meaning as they have moved from clinical to common, typically negative,
usage. Invoked as slurs, these terms draw a significant measure of their derogatory
power from their association with the legitimacy of the clinical judgement practised by
psychiatrists.
Participants’ narratives confirmed the findings of Reavley and Jorm (2011), that there is
still widespread stigma in the community against people with mental illness. Moreover,
this stigma remains more pronounced against different diagnoses, with people with
schizophrenia being the most stigmatised, borne out by the psychiatrist’s reported
comment to Shirley’s mother that, ‘you’re lucky she doesn’t have schizophrenia’.
In this context, Ian’s suggestion that if schizophrenia were called a ‘nice name…like
brainiac’, then people would hold less stigmatising attitudes toward people with the
diagnosis, while made humorously, actually calls up serious arguments about the
influence of language. The issue of whether diagnostic terms are in themselves
stigmatising arose several times in participants’ reports, including Ian’s dislike of the
prefix ‘schizo’. Bleuler’s (1911/1950) attempt to distance schizophrenia from the
previous stigmatising association with dementia (Fusar-Poli & Politi 2008) itself yielded
to stigma one hundred years later, as evidenced by calls for the term to be abandoned
altogether (Kingdon et al. 2007).
Shirley expressed a more sceptical view, ‘that stigma follows you round whatever
language you use’. Her assessment that stigma operates regardless of the terms used
aligns with the findings reported by Tranulis and colleagues (2013), that the benefits of
changing the name for conditions is attenuated over time as society comes to
understand the correlation with the previous term.
Stigma has been comprehensively challenged in recent decades with a number of
previously marginalised social groups collectively asserting their rights for recognition,
and rejecting the status quo wherein their perceived attributes were used to justify
Chapter 6 Discussion
152
lesser rights. The statement ‘The personal is political’ (Hanisch 2006), popularised
within the feminist movement, was a catch-cry of people advocating for change by
foregrounding the salient elements that had previously been used to justify oppression
of certain groups. This was a converse approach to the earlier strategy of passing, and
it was members of many of the same groups who had once sought social advantage by
masking their stigmatised attributes who now adopted the reverse strategy, under the
banner of ‘identity politics’.
A powerful strategy used by these groups has been to take the metonymic identifying
terms used by others about them, and re-appropriate them, making them a badge of
pride. ‘Nigger’, ‘faggot’, and ‘slut’ are all words that have been used in this way. It is
always a contested strategy, with some members of the groups in question rejecting
the terms as perennially harmful. Mad pride is the most visible group within the mental
health consumer movement who have utilised these strategies, reclaiming the term
‘mad’ in preference to diagnostic terms (Dellar, Curtis & Leslie 2000). While several
participants challenged negative stereotypes about their capacity or dangerousness,
none of the participants in the current study adopted this particular stance, and several
commented that they themselves don’t loosely use terms such as ‘crazy’, and don’t like
it when others do. This demonstrates deliberate language choices, used to distance
themselves from stigmatising positions.
SELF-STIGMA
A specific form of stigma is self-stigma, or internalised stigma. This occurs when
people regard themselves as members of a group and endorse stigmatising attitudes
toward the group that are held more broadly in the community. Ethan observed that,
‘Insight is a double-edged sword’, that is, it can help in gaining control of your
symptoms, but, if you hold negative ideas about mental illness, it can also result in low
self-esteem and poor mental health outcomes.
The strongest example of the effects of self-stigma was Ian’s reported non-adherence
to treatment for schizophrenia for twenty years, with multiple involuntary admissions to
hospital and a great deal of suffering. Subsequent to his diagnosis being reformulated
from schizophrenia to bipolar disorder, he reported improved adherence, even though
this involved him taking the same medication he was previously prescribed, as well as
an additional mood-stabilising drug. The result of this improved adherence was an
improvement in Ian’s control of his symptoms. This demonstrates the dual operation of
literal and metonymic language and thinking (Littlemore 2015). On the literal level, the
re-diagnosis and addition of a new pharmacological agent contributed to the
Chapter 6 Discussion
153
improvement in his mental state. But in Ian’s telling, the metonymic shift in meaning
that he attributed from one diagnosis to the other, that is, what each stood for in
relation to his sense of identity, was integrally tied to his willingness to adhere to the
treatment, and consequently resume better control of his illness.
Lysaker and colleagues (2006) explored self-stigma in a study in a large psychiatric
hospital. They found that a significant proportion of people with schizophrenia
expressed good insight into the fact that they had a mental illness, but also internalised
stigmatising attitudes about mental illness. For this group, outcomes in terms of their
mental illness were worse than for those who either had minimal insight, or those
whose insight was not accompanied by negative attitudes toward mental illness.
Corrigan and colleagues (2013), drawing on research that demonstrates the benefits
that accrue to LGBTI people when they ‘come out of the closet’, that is, disclose their
previously invisible sexuality, developed a program using a similar technique to support
people who live with mental illness to weigh up the costs and benefits of disclosing in
various contexts. The program was subsequently tested in a randomised control trial
(Rüsch et al. 2014), with participants reporting reduction in the stress related to selfstigma.
STIGMA FROM HEALTH WORKERS
Participants reported their perceptions of stigmatising attitudes held by health workers
in both general and mental health settings. Shirley reported an incident when her acute
respiratory distress was dismissed as delusional. Ian reported that after a suicide
attempt, staff stitched his wounds, but did not address his mental distress. Yves
reported anti-cholinergic medication being withheld by clinicians on the basis of his
past history of misuse of the drug, even though they acknowledged the clinical
evidence of acute dyskinesia (a medication side-effect for which anticholinergic
medication is prescribed). Ellen reported that, once having been scheduled as mentally
ill, she was treated as if had also lost her intelligence, and was subsequently excluded
from decisions about her treatment.
Participants reported they did not discuss these perceptions of stigmatising treatment
directly with clinicians. They cited the primary reason for this reluctance was the power
imbalance between clinicians and people who live with mental illness. This imbalance
is not restricted to the power granted to clinicians through legislation to detain and treat
people coercively if they meet certain criteria, though this power had been utilised in
the lives of nine out of the ten participants.
Chapter 6 Discussion
154
The power imbalance in the historical practice of psychiatry has been documented by
Foucault (1980). Slade (2009) noted that this imbalance persists in current clinical
practice in the relative weight given to knowledge held by people who live with mental
illness and clinicians in consultations concerning treatment decisions. The imbalance
also influences current research paradigms which privilege positivist models of
knowledge over evidence from the lived experience of people who live with mental
illness (Slade, 2009).
Goffman (1963, p. 76) identified a specific exercise of power that he termed ‘phantom
acceptance’. Shirley’s reported experience with her treating psychiatrist illustrated this
concept. Her psychiatrist supported her engagement as a consumer advocate, and
displayed a health promotion brochure Shirley had co-authored in the clinic waiting
room, granting recognition to her advocacy work, or ‘phantom normalcy’ (Goffman
1963, p. 76). However, when Shirley requested that her psychiatrist implement the
intervention in her own health care, this was dismissed with the phrase ‘I don’t have
time’. The support for Shirley’s collaborative initiative was performed in public, but
disallowed within the private therapeutic engagement, revealing the ‘acceptance’ as
‘phantom’.
The psychiatrist’s reported refusal to provide the requested care was not explicitly
stated as a refusal, but expressed metonymically as ‘I don’t have time’, where time is
made to stand for intent. This is an example of how metonymic under-specification is
strategically used (Littlemore 2015). A request is refused but the substitution of time for
intent allows the speaker to deny that refusal is the intended inference, in turn creating
the suggestion that any perceived slight is imagined (Goffman 1963). Eggins and Slade
(1997) have pointed out a similar role for humour in workplace conversations, where
humour can be used to form an in-group whose members can share jokes that isolate
others, but subsequently deny this intent if it is explicitly questioned.
Participants also reported poor access to physical health care once their mental illness
is disclosed, which they experienced as stigmatising. Shirley used the term ‘diagnostic
overshadowing’ to report her experience of poor treatment by a triage nurse in an
emergency department. This term was used in a study in which emergency department
clinicians reported multiple cases of misdiagnosis or delayed evaluation and treatment
of people who live with mental illness (Shefer et al. 2014).This process has been
reported by people who live with mental illness accessing emergency departments
throughout Australia (Australian Commission on Safety and Quality in Health Care
2014).
Chapter 6 Discussion
155
Such treatment can reflect the problem of stigmatising attitudes held by clinicians
toward people who live with mental illness (Ross & Goldner 2009). When these
attitudes result in the delivery of sub-optimal care, the function of stigma has moved
from prejudice to discrimination (Thornicroft et al. 2007). Goffman (1963) described
how identification of a stigmatising attribute in a person enabled ‘normals’ to ignore any
claims the person may make based on other attributes. This can be extended to
include claims for comprehensive health assessment and care. Hill (2010) has reported
on neurological evidence of reactions of moral repugnance by clinicians, but also
demonstrated that these deeply felt reactions are amenable to change with training and
support.
OTHER METONYMIC INFLUENCES ON THE EXPERIENCE OF
TREATMENT
Experiences of sub-optimal treatment were not always perceived to be motivated by
stigma from health workers. Participants reported experiences that were indicative of
partial attention rather than prejudice. Shirley observed that the devastating effects of
the depressive phases of bipolar disorder were frequently overlooked by health
workers, who focused instead on people’s presentation in manic states, when they
were ‘bothering the public’. Ethan, Nathan and Oliver all reported the experience of
their diagnosis not being clearly explained to them, even as treatment based on that
diagnosis was being initiated. They reported experiencing this clinical reticence as
more disturbing than receiving the diagnosis.
These accounts echo similar experiences reported by participants in Barham and
Hayward’s (1991) study, conducted over 20 years ago. The authors observed that:
Many of our participants described how they received little or no guidance in
tackling the meanings of schizophrenia, and were left to cope with the cultural
burden of the diagnosis as best they could (Barham & Hayward 1991, p. 21).
They are also consistent with contemporary data collected in the Obsessive Hope
Disorder report, where 20% of their respondents indicated that diagnosis was not
discussed by their treating teams (Mendoza et al. 2013), and in a study by Cleary and
colleagues (2010) in which 76% of the respondents rated being provided with a
diagnosis as ‘very important’. The confusion that people are left with when clinicians do
not provide clear information was also reported in the Scottish Recovery Network
project:
Chapter 6 Discussion
156
Some narrators felt let down by the professional services delivering their
diagnoses. They felt that diagnoses were sometimes suppressed by
professionals only to be given at a later date without information or support
leaving them to feel disempowered about their own health (Brown &
Kandirikirira 2007, p. 24).
While this process can be seen to be different from metonymic over-determination, it
stems from a similar conceptualisation of the person, by dint of their diagnosis, not
having capacity to participate in treatment decisions. In cognitive linguistic terms, this is
effectively a combination of metonymy with ellipsis (Al-Sharafi 2004); clinicians
metonymically deduce that the person lacks capacity to cope with the diagnosis of
mental illness, and so they elide reference to this in their discussions.
An explanation stems from the way in which clinicians routinely practise cognitive
shorthand in making assessments, selecting what they view as salient information.
Making a diagnosis involves trained use of focus, and setting aside extraneous
elements. Oliver commented that in his experience, this process was hastened by
resource issues. If a person’s history of mental illness is selected as the salient
element, a clinician may focus on this, and miss, or defer attention to other elements.
This is not necessarily a question of stigma, but of the cognitive operation of attention,
where foregrounding of an element necessarily forces other possible elements into the
background (Talmy 2007). James (1890/1950) noted how attending to one attribute
involves ignoring other available evidence. Problems arise when this practice entails
habitual foregrounding or backgrounding of mental illness, rather than clear focus on
the current presenting problem. Patel and colleagues (2015) have investigated the role
of cognition in both error generation and error recovery in health care, and suggested
ways in which this knowledge can be used to improve safety. Metonymy is a cognitive
process that frequently goes unremarked yet, as participants’ narratives demonstrated,
can be just such a source of cognitive error.
While inattention can be explained cognitively, there is also evidence that some
clinicians deliberately avoid responding to patient’s cues, particularly when these are
emotional in nature (Uitterhoeve et al. 2009), or expressed in figurative language.
Lancely and Clark (2013) observed the practice of some nurses who responded
effectively to the emotional needs of people in treatment for cancer, but others who
evaded this engagement, seemingly ‘excused’ by the fact that people often expressed
their fears in metaphoric terms, which the nurse could claim not to have understood.
Chapter 6 Discussion
157
This aligns with participants’ reports of believing clinicians were too frightened to talk to
them about their experience.
Participants reported experiencing clinicians underestimating their capacity to make
decisions once they had been diagnosed with mental illness, or made involuntary
patients under mental health legislation. These practices have been addressed in
changes introduced into the two most recently amended mental health acts (Mental
Health Act (Victoria) 2014; Mental Health Act (Western Australia) 2014), which include
provisions requiring clinicians to demonstrate they have assessed the person’s
capacity to give informed consent before every treatment decision.
The influence of metonymy is also evident in participants’ reports of good experiences
of treatment. Participants reported on the positive influence of careful language choices
by clinicians, in particular, language that positioned both the person providing and the
person receiving care as people engaged in communication. They also noted the
beneficial effect when clinicians used unfamiliar language such as psychiatric
diagnoses carefully, ensured the person receiving the information understood, and
were open about what they were thinking in terms of the relation of participants’
reported experiences to possible diagnoses. These are the kinds of practices identified
in a recent qualitative study published on the topic of the needs of people receiving
serious mental health diagnoses (Milton & Mullan 2014). This study noted that people
wanted information at the time of diagnosis, but wanted that information to be tailored
to their individual understanding. It also noted that individuals wanted assistance to
manage their own self-stigmatising ideas, based on previous misconceptions of mental
illness. Commenting on how this is handled, one of their participants observed, “Some
[clinicians] would treat the illness, some would treat the person, and the ones that
treated the person would have better outcomes” (Milton & Mullan 2014, p. 462). This
reiterates the concept that metonymically foregrounding the person before the
diagnosis contributes to better treatment.
Participants’ narratives in the current study further suggested that it is not just people
who live with mental illness who are viewed as constrained by their position in the
encounter, but that health workers can be equally constrained by their roles. They
described positive encounters with clinicians, but these were presented as outside their
expectations. This indicates a mismatch between contemporary portrayals of holistic
practice, and people’s actual experience of care. Peplau (1952/1988, p. 70) addressed
the possibility of therapeutic interactions benefiting both participants fifty years ago:
Chapter 6 Discussion
158
Nurses often symbolize nonrational roles to patients, that is, they stand for but
are not mothers of patients, and they take on these roles at the same time
helping the patient to clarify his preconceptions and to become aware of the
nurse as a person in her own right.
The symbolic role Peplau (1952/1988) posits for nurses is metonymic; they ‘stand for’
but are not the patients mothers, and in performing this role for the patient, they are
also enacting selfhood for themselves. Walker (1994, p. 165) has suggested that
narrative contributes to this work:
Nurses lives then, can be theorized as being inextricably woven out of narrative
structures which work to impart coherence and order…on the ‘rough and
tumble’ of everyday life in practice.
This underlines the idea that nurses have a need to make meaning of their experiences
in therapeutic encounters as well as the people they care for. Following Menzies Lyth
(1959/1988), this kind of nursing can only occur within a health care system that
enables the opportunity for workers and people receiving care to form therapeutic
attachments. Participants reported their experience of considerable benefits when they
encountered therapeutic engagement, and confusion and increased distress when it
was absent.
METONYMY AND THE EXPERIENCE OF SYMPTOMS
Metonymy, as both a conceptual and a linguistic process, influenced participants’
accounts of their experience of symptoms. They reported delusional beliefs that
demonstrated inferences linked to metonymic sources. In some cases, the metonymic
link between a sign and the inference drawn from it was relatively straightforward.
Nathan inferred that he was expected to die soon when he was placed in hospital near
old people. Ian’s guilt about illicit substance use was manifested by experiencing a
voice telling him he was going to go to hell. Shirley’s description of London as, ‘quite a
good place to sort of flip out, because everything actually is sort of connected’,
highlights the fact that the capacity to infer specific meanings from locations is available
to anyone. The development of other reported delusions involved a further meaningmaking step. Ethan’s real sense of powerlessness was transferred into the belief he
was a political prisoner. Shirley’s fear of harm coming to those close to her was tied to
the tension she experienced as an identified critic of drug companies. These can be
seen as examples of metonymic chaining (Brdar-Szabo & Brdar 2011).
Chapter 6 Discussion
159
These reports align with the work of Rhodes and Jakes (2004), psychologists who
suggested a metonymic motivation for a delusion reported by a study participant with
schizophrenia. Their suggestion was subsequently taken up by the cognitive linguist
Littlemore (2015), who identified further potential metonymically motivated delusions in
Rhodes and Jakes (2004) original data set. In both cases, the metonymic links the
authors draw are conceptual and not linguistic. Similarly, the cognitive links in the
participants’ narratives are between signs perceived in the world and internal
inferences they subsequently draw. This is different to inferences being drawn through
the use of metonymic language, as often the links have not yet been expressed
linguistically.
With the research methods currently available, theories about possible metonymic
motivation for delusional beliefs cannot be confirmed. This is consistent with the
inexact nature of following thought patterns, which Hobbes (1650/2011) described.
Psycholinguists have used neuroimaging techniques to track the activation of specific
brain regions during language performance, in order to identify correlations between
brain activity and clinically demonstrated thought disorder, but they concede that their
findings, though suggestive, remain speculative (Kuperberg, Kreher & Ditman 2009).
Empirically, the participants in the current study were unable to confirm metonymic
motivations for delusional beliefs with any more certainty than the researcher.
The reporting of symptoms is a key point of intersection between the two research
questions in the current study. Descriptions of delusional beliefs suggest the influence
of metonymy, in its conceptual sense, on the experience of people who live with mental
illness. However, these experiences are not necessarily expressed in metonymic
language. Conversely, when describing symptoms more broadly, including symptoms
such as hallucinations that are not linked to conceptual metonymy, participants used a
range of language practices, including figurative language, in order to report
experiences that they described as difficult to convey. Thus, the influence of metonymy
in their speech production was tied to their capacity to give voice to their experiences.
When asked how he described his symptoms, Nathan chose to use the exact clinical
diagnosis he had been given in order to describe his experience, ‘depression with
psychotic features’. Tanya reported using different terms depending on the context in
which she was talking, though she inverted the usual order, using ‘schizophrenia’ with
acquaintances, but less specific terms such as ‘mental illness’ when conferring with
clinicians. Michael used metaphor and metonymy, describing the ‘noise musicians’
Chapter 6 Discussion
160
inside his head. Shirley and Ian described how difficult it is to articulate the experience
of symptoms. Ethan used the metaphor ‘descent’.
The metaphor of mental illness as ‘descent’ was a concept given currency by the work
of R.D Laing (1960) and other early psychiatrists who were active in the ‘antipsychiatry’ movement in the 1970s, most notably given first-person voice by Mary
Barnes (1971). There are parallels with the frequently figurative terms used in clinical
discourse, such as ‘flight of ideas’ (Akiskal 2009), which serves as an antonym for
descent. The cognitive links between the metaphoric term and the embodied
experience of negative emotions have been described by cognitive linguists (Kovecses
2006; Lakoff 1987), who have noted the bodily experiences such as lethargy and
drooping posture that align with, and possibly underpin the concept of depression as
descent.
Participants used metaphor when they were describing their experience of psychosis.
In response to their sense of the ineffability of the experience, they spoke of what it
was like, rather than what it actually was. In contrast, when speaking of the functional
and pragmatic effects of the illness in general, participants strategically used
metonymic language to mark what it ‘stood for’ in their lives. Thus, Michael reported
that ‘“recovery from schizophrenia” sounds much nicer’ [than schizophrenia]. Shirley
spoke of limiting the conceptual power of psychosis by referring to it as ‘scary thoughts’
and reported this had a pragmatic effect in her recovery from her most recent psychotic
episode. Michael was thus exploiting the foregrounding aspect of metonymy (Talmy
2007), highlighting the recovery over the diagnosis while keeping both visible. Shirley
was using the ‘stand for’ aspect (Feyaerts 2000), displacing the clinical ‘psychosis’ with
the more quotidian ‘scary thoughts’, using indirect reference to render the experience
more manageable. Participants’ use of metonymic language is explored further in the
following section.
The influence of metonymy on the speech production and
speech comprehension of people who live with mental
illness
The answer to the second research question, ‘How does metonymy influence the
speech production and speech comprehension of people who live with mental illness?’
comes in two parts. In the first instance, there was some evidence of problems in the
production and comprehension of metonymic language. While the speech production of
participants minimally evinced unusual word choices, consistent with the ‘metonymic
Chapter 6 Discussion
161
speech’ identified by earlier researchers (Cameron 1944; Goldstein 1944), participants
did demonstrate and report difficulties with comprehension of speech, particularly
figurative speech, in line with earlier research findings (see, e.g. Chapman 1960;
Kuperberg, Kreher & Ditman 2009).
In the second instance, participants demonstrated the capacity to both produce
metonymic references, and accurately process metonymic inferences spoken by
others, consistent with cognitive linguistic understanding of the ubiquity of the figure in
everyday speech (Lakoff 1987; Littlemore 2015).
METONYMIC SPEECH
I detected rare examples in participants’ narratives of the obscure indirect referencing
defined as metonymic speech by Cameron (1944) with Tanya’s reference to ‘the lionhead pond’ being the most idiosyncratic and potentially opaque example. This finding is
consistent with studies by Andreasen and colleagues (Andreasen 1979a; Andreasen &
Grove 1986), that included metonymy under the term ‘word approximations’, but
uncovered little evidence of its use in their data. It can be interpreted in light of three
issues. First, Cameron’s studies (1938; 1944) were undertaken prior to the discovery of
anti-psychotic medications, and so the people with mental illness that he engaged with
remained in psychotic states for longer periods than is common today, potentially
leading to more acutely disordered language production. Second, the way that
metonymy is conceived has undergone significant changes in the last 70 years, from
being restricted to a figure of rhetoric to being understood as contributing to the way in
which experience is cognitively understood and communicated (Lakoff & Johnson
1980; Littlemore 2015). This has led to broader recognition of the operation of
metonymic inferencing. Third, data in this study were collected through narrative
interviews and so participants had control of the context they established; as the
researcher, I was primed to infer the salient meanings for terms by the preceding
content of their stories, and therefore less likely to capture the kind of irregularities
noted in previous context-free studies.
COMPREHENSION OF METONYMY
Participants demonstrated or reported difficulties in speech comprehension in relation
to the meaning of words, particularly when these meanings changed in different
contexts. Michael’s remark that, ‘often I’m caught out, because in this short period
something’s changed, and I haven’t really kept up with the change’ resonates with
Cameron’s description of trying to understand one of his research participants: ‘One
Chapter 6 Discussion
162
goes along for a bit all right, but then begins to slip behind and miss the meaning’
(Cameron 1944, p. 54).
Responses to the ‘quick quiz’ question, ‘Would you like another glass?’ demonstrated
the range of influence metonymy has on the experience of people who live with mental
illness. For the majority, processing the intended non-literal meaning, that is, the
contents of the glass rather than the container itself, posed no problem. While the
‘quick quiz’ revealed comprehension difficulties for some participants, it also evinced
others’ capacity to chain metonymies together, to infer meaning that may be prompted
by context, but incorporated world knowledge external to language (Brdar & BrdarSzabo 2003). This was best captured in Shirley’s response, ‘Well, I’d be thinking
Chardonnay’.
It was the intensity with which the ambiguity of the non-literal request was noted that
distinguishes the responses of participants. Michael responded with tangentiality,
talking about his ideas about language in place of answering the question. Ethan
inferred a threat from the possibility that the glass be used to ‘glass’ someone,
suggesting an atypical inferential schema consistent with paranoid ideation, but also
the activation of mental concepts through semantic priming (Kuperberg & Caplan
2003). Even more pronounced was the disturbance experienced by Matthew in
response to not being able to comprehend the intended inference of the question,
accompanied by the social unease engendered by this inability.
These findings are evidence of the language comprehension challenges faced by some
people with mental illness, and previously observed by researchers (Chapman 1960;
Kuperberg & Caplan 2003). Participants reported on the impact of these
communication difficulties. Michael and Tanya each spontaneously reported that they
have simply stopped talking at times, and others reported situations where they have
been reluctant to engage in social interactions because of the concerns in relation to
their capacity to communicate.
The issue of people not talking has broader implications, as Goffman (1963) flagged. If
people with lived experience of mental health issues stop talking, not because they
want to, but because they cannot keep up with conventional conversational practices,
this increases their social isolation and inhibits their interactions with others. These
elements can lead to further deterioration in their social functioning and mental state
(Bowie & Harvey 2008). Moreover, the contribution they can make to community and
social life is lost (Barham & Hayward 1991).
Chapter 6 Discussion
163
LINGUISTIC AND CONCEPTUAL METONYMY IN PARTICIPANTS’
‘EVERYDAY’ LANGUAGE
The study provides information about the overall use of language by people who live
with mental illness, rather than just evidence of impaired use. As members of the
broader speech community, participants demonstrated capacity to produce and
comprehend metonymic speech, both in their own free narratives, and in response to
questions from the interviewer. Metonymy appeared in combination with other
figurative language in blends and idioms, in phrases that are so common their
metonymic origin is hard to discern, and in novel and original ways.
Participants demonstrated sophisticated skills in using metonymic language to create
intended inferences. Oliver combined ellipsis and metonymy (Al-Sharafi 2004) to report
his dissatisfaction with the care he received, while graciously avoiding attributing blame
to specific individuals. Ian used a metonymic reference, ‘the boys would be there’, to
counter an initial joking question as to whether he was a gelding when he recounted a
story of imagining himself as a prize-winning horse. Veale, Feyaerts and Brone (2006)
have demonstrated how metonymy facilitates this style of humour, as the speaker can
show they have understood another’s inference, and ‘trump’ it, by using a further
indirect reference. Shirley’s strategic use of the metonymic ‘pen on the prescriber’s
pad’ demonstrated adept use of media-friendly, readily inferred language, consistent
with the renewed interested in metonymy’s effectiveness in the rhetorical practice of
persuasion (Littlemore 2015). Each of these examples showed that participants
exercised these linguistic skills in the performance of social interactions: Oliver allowed
others to save face; Ian bonded with a stranger through humour and Shirley exploited
the constraints imposed by media to represent clinical accountability.
Rochester and Martin (1979, p. 24) identified a problem with the traditional approach to
research on language use by people who live with mental illness:
The aim has been to capture, not the distributional properties of the corpus, but
its deviant features. That is, rather than describing the utterances as a whole,
investigators of schizophrenic speech have attempted to characterize those
features of the corpus that differ from normal. In effect, the effort has been to
describe the failures rather than the overall performance of the schizophrenic
speaker.
This focus characterises the contemporary approach taken by psycholinguists, as they
themselves note (Kuperberg & Caplan 2003). By presenting the full breadth of
Chapter 6 Discussion
164
participants’ experience of metonymic and pragmatic inferencing, the current study
contributes to broadening our understanding of language as it is experienced by people
who live with mental illness.
NARRATIVE
The initial interviews with participants sought to elicit their narratives in their own words.
Participants deployed a significant range of narrative skills. Ian and Ellen were adept at
embedding smaller narratives within larger ones. Tanya’s narrative included the
occasional unprompted insertion of heavily detailed vignettes. Michael engaged in
discursive play with the narrative metaphors of journey and myth. Shirley, who has
narrated her experience to a variety of audiences, used the term ‘patient porn’ to
describe a particular relation between the story of the person who lives with mental
illness and the story’s consumption by others. Ethan described the ongoing evolution of
his story and the contribution of this to his recovery.
Oliver, Ellen, Stan and Shirley frequently combined reported speech and direct speech
in their narratives of past experiences, even when it was not feasible that they could
recall exact words used in conversations held years ago. In literary terms, such a
practice constructs verisimilitude, that is, it creates the appearance of a ‘true’ narrating
of the episode (Bruner 1997). The ‘unreliable narrator’ (Lodge 1992) has been a trope
in literature long before qualitative researchers cautioned against taking informant’s
words completely at face-value (Hammersley 2008). This is not to say that participants
are being deliberately misleading when they adopt such techniques in telling their
stories. Rather, it is to recognise that they utilise conventionalised strategies to create
meaning.
Michael’s account of coming to the recognition that he had a mental illness in the midst
of his tenth involuntary hospitalisation is an example of another narrative technique,
beginning a story in media res, that is, in the middle of the action, with the beginning
not explained. This was an example of a participant’s narrative extending beyond the
straightforward chronological order identified in traditional narrative models (Labov &
Waletzky 1967/1997).
Narrative as a metaphor for life underpins narrative practice (White 2007). Moreover,
some of the strategies participants adopted in reconstructing their experiences in
narrative were metonymic, with specific elements of their past selected for
foregrounding. James (1890/1950, p. 571) observed:
Chapter 6 Discussion
165
In no revival of a past experience are all the items of our thought equally
operative in determining what the next thought shall be. Always some ingredient
is prepotent over the rest. Its special suggestions or associations in this case
will often be different from those which it has in common with the whole group
of items; and its tendency to awaken these outlying associates will deflect the
path of our revery. Just as in the original sensible experience our attention
focalized itself upon a few of the impressions of the scene before us, so here in
the reproduction of those impressions an equal partiality is shown, and some
items are emphasized above the rest. What these items shall be is, in most
cases of spontaneous revery, hard to determine beforehand.
At times, eliding certain events was a deliberate choice, with participants collaborating
with the interviewer to foreclose discussion of traumatic events. However, in line with
James (1890/1950), participants frequently demonstrated spontaneous shifts midnarrative, as they recalled and recounted events prompted by the act of narrating itself.
Running counter to the frequent indeterminacy of both narrative and memory is a broad
cultural constraint on people to provide coherent accounts of themselves: ‘Those who
are in a strict sense not capable of biographically presenting themselves create
interactional disturbances’ (Fischer-Rosenthal 2000, p. 116). This narrative demand
accompanies a pressure to present a coherent self to others. It is arguable whether the
demand extends to actually having a coherent self, but certainly people are subject to,
and aware of an implicit demand to maintain a coherent, contained self in their
interactions with others. This also links to the issue of visibility addressed earlier in the
chapter.
Ethan reported that the act of narrating provided cohesion to his sense of his
experience. This echoes Stuhlmiller’s (2001, p. 65) concept of narrative as an
organising process:
Not only do narratives metaphorically create categories for interpreting events,
but also they bind people and events into some intelligible pattern.
Stuhlmiller (2001) highlights how the metaphor of life as a narrative supports its use as
a strategy for meaning-making. For Ethan, who described the experience of mental
illness as ‘disintegration’, it is the capacity of narrative to ‘bind events… into some
intelligible pattern’ (Stuhlmiller 2001, p. 65) that enables him to view his life as other
than ‘a slag heap’. The structuring of his life in narrative terms initially creates a sense
of cohesion in relation to his experience. Communicating this narrative in turn
Chapter 6 Discussion
166
contributes to rendering the experience of mental illness intelligible to other people. In a
third, recursive step, the ‘evolving’ retelling in turn builds the narrative that sustains him
in his recovery. This ongoing process enacts the therapeutic effect of narrative practice
(White 2007) in its purest form, dispensing with the therapist entirely.
Shirley’s use of the term ‘patient porn’ draws on a contemporary critique of the
potentially co-optive uses of the personal narratives of people who live with mental
illness by mental health services and researchers (Costa et al. 2012). In light of this,
and echoing Ethan’s invocation of the transforming power of narrative, she reported
that she deliberately used the constraining form of the narrative interview requested
within the current study to explore aspects of her own experience that she had not
previously disclosed.
Contribution to cognitive linguistics
While not explicitly setting out to do so, the study contributes to the literature on
cognitive linguistics. Cognitive linguistics has been criticised as relying too much on
examples that are constructed by researchers, rather than drawn from actual speech
(Gibbs 2007b). This study contributes to the recent development of ‘corpus-based’
cognitive linguistic research, using examples of actual speech to understand the
cognitive processes occurring. Whereas many of these corpus studies draw on existing
bodies of language (see, e.g. Markert & Nissim 2003), this study constructs a new body
of data, 400 pages of narrative interviews with people who live with mental illness.
From a cognitive linguistic perspective the results of the study demonstrate that people
with serious mental illness have the capacity to both construct and interpret
sophisticated figurative language in online speech. It demonstrates that while
participants do indeed produce metonymic speech, this is consistent with conceptual
metonymy as described by cognitive linguists (Lakoff & Johnson 1980; Littlemore 2015)
rather than the ‘metonymic distortion’ identified by psychiatrists early in the twentieth
century (Cameron 1944; Goldstein 1944). It also demonstrates that some participants
experience difficulties in comprehending figurative language, consistent with findings
reported by earlier researchers of research participants being unable to follow shifts in
inferences attached to words used in specific contexts (Chapman 1960; Rochester &
Martin 1979). This is linked to the cognitive linguistic principle that words trigger an
encyclopaedic response rather than a dictionary one (Langacker 2007). Ironically,
Michael explicitly invoked this concept when asked if he sometimes missed other
people’s meanings in conversation, ‘Well, that’s what the dictionary’s for’.
Chapter 6 Discussion
167
Littlemore (2015) has suggested that greater attention to metonymy may improve the
teaching of English as a second language, psychotherapy, and political discourse,
through recognition of the inferences that are created, intentionally or otherwise, by the
use of metonymic ‘hidden shortcuts’. The current study suggests that clarifying the
influence of metonymic pragmatic inferences in the language they encounter may
improve the experience of people who live with mental illness.
Limitations of the study
This is a qualitative, descriptive study, designed to explore the experience of language
of people who live with mental illness. Analysis of the use of language is observational
and descriptive. As noted, every person’s language evinces different characteristic
patterns (Jakobson 1956/1987). The study does not provide analysis of the language
use of a comparison group of people who do not have diagnoses of mental illness. As
someone who has worked as a mental health nurse for more than fifteen years, it is
possible that long-term interaction has reduced my capacity to notice speech produced
by people with mental illness as unusual.
The study method included interviews which were audio recorded and transcribed.
Subsequent analysis was restricted to analysis of the text of the interviews. Andreasen
(1979a) has observed that non-verbal data can be lost in such a process, and
suggested that combining audio recording with video recording may attenuate this loss,
without however, outlining how these data could be analysed. The analysis in this study
is limited to transcripts of audio recordings. While some notations were made, including
significant pauses, laughter, and gestures used intentionally to suggest meanings
(pointing at the recording device, circling arms around head), there is not an analysis of
non-verbal communication. This is for a number of reasons. The research explicitly
addresses the creation of meaning and inference through language, and therefore it is
appropriate to consider participants’ language. There are reliable methods for
analysing language as text, and text remains available for interested parties to refer to,
to check analytic decisions.
There was a potential for self-selection bias among the participants, that is, only
participants who had an interest in language would agree to participate in the study. A
number of participants did express significant interest in language, both language use
in general, and language use around mental illness in particular. For a number of
participants, this interest is professional, as they are people whose work life has
involved language practices including narration or ‘story-telling’, writing and training.
Chapter 6 Discussion
168
All participants were currently taking medication. Anti-psychotic medication and other
neuroleptics, including mood-stabilising medications, have side-effects which include
influencing cognition. At a basic level, people can experience some sedation, and also
a slowing of the speed of their thought processes. This may in turn make changes to
the information collected in interviews. The aim of the study was to explore the
influence of language on the experience of people who live with mental illness, rather
than language during acute psychotic episodes, and most of the participants credited
medication as a significant factor in their recovery. Therefore it made sense to interview
people in the recovery phase, incorporating the potential effects of prescribed
medications.
Implications for further research
This study has investigated the influence of metonymy on the experience of people
who live with mental illness, using their narratives and language, but reported from the
perspective of a single researcher with a background in mental health nursing and
policy. The practice of co-research (Speedy 2005), where investigators and participants
work together as partners rather than researcher and research subjects, is gaining
momentum in the mental health context (Costa et al. 2012). This approach would
potentially be effective for further research exploring the construction and negotiation of
meaning and inference, which is necessarily a joint process.
The current study included information about how people who live with mental illness
experience the language used by mental health workers. Complementary research
could investigate mental health workers’ thoughts about the strategic use of diagnostic
language, their understanding of the complexities differentiating all language users
from each other, and their observations of the influence of the language they use with
people who consult them.
Cross-language research in cognitive linguistics has revealed that there are some
similarities and some differences in use and interpretation of figurative language in
different languages (Brdar-Szabo & Brdar 2012; Radden & Seto 2003). The limited
research on the experience of people from culturally and linguistically diverse
backgrounds who live with mental illness indicates that they experience greater
difficulties in their communications with mental health workers (Minas et al. 2013).
Research that mapped the influence of metonymy across languages in the mental
health context may help to address these issues.
Chapter 6 Discussion
169
Participants reported that some of the most effective encounters they have had have
been with other people who live with mental illness. Beyond the benefits of support,
they have identified the sharing of specific techniques is effective in their own
management of their experience. The research on peer support workers has been
criticised as not providing rigorous enough evidence (Lloyd-Evans et al. 2014).
Notwithstanding this concern, the employment of peer support workers, both as direct
service providers, but also as trainers for undergraduate and existing practitioners is
increasing (Repper & Carter 2011). Reflecting participants’ views on what actually
works for them, consideration may be given to research that navigates between
establishing rigour and supporting innovation in this area.
The current study did not produce evidence about the effectiveness of explicitly
drawing attention to metonymic language and thinking in therapeutic work, though this
approach formed part of my previous clinical work. The study has demonstrated the
prevalence and influence of metonymy on people’s experience. Participants
demonstrated sophisticated understanding of language they encountered, and
competency in using language themselves. When metonymic links were made explicit
by the researcher, participants frequently expressed that they grasped the ideas, and
could trace their influence on their thinking. In doing this, they also demonstrated
recognition of how metonymic concepts underpin significant topics of concern, such as
identity and stigma, opening the possibility of exploring its therapeutic potential in future
research.
Implications for practice
Participants reported encountering stigma from health workers. They reported that
stigma was conveyed through language, through attitude, and also through the delivery
of sub-optimal care, for both mental and physical problems. Stigma against people who
live with mental illness has been a subject of concern for more than fifty years
(Goffman 1963), yet it persists (Hill 2010; Reavley & Jorm 2011; Ross & Goldner
2009), with participants reporting stigma-related actions by health workers occurring
from the 1970s through to 2014. This study supports the call of Thornicroft and
colleagues (2007) for attention to shift to acts of discrimination rather than further
surveys of attitudes. Health service organisations have responsibilities under
governance standards to ensure services are provided to all without discrimination
(Australian Commission on Safety and Quality in Health Care 2014), and could use
performance management systems, supported by education and adequate resources,
Chapter 6 Discussion
170
to enable health workers to provide health care without stigma to people who live with
mental illness.
Participants reported encounters with mental health workers that were not stigmatising,
but nonetheless frustrating. These were frequently related to language. Influential
resources used in making diagnoses (American Psychiatric Association 2013), and
preparing people to work in mental health (Sadock 2009), present limited views of
communication, with the person’s language subject to scrutiny, but the health worker’s
presumed to be transparent and effective. This is not consistent with participants’
reports of their communications with health workers. Encouragement for clinicians to
understand their own cognitive and communicative processes is likely to improve the
experience of the people they encounter, and the safety of the care that is delivered
(Patel, Kannampallil & Shortliffe 2015). Following Zola (1993) and others, clinicians
have learned to refer to people without nominalising the diagnosis, and following
Vygotsky’s theory of knowledge development, the subsequent ‘zone of proximal
development’ (1978, p. 86), would be to develop understanding of the concepts
underpinning this language use. Recommendations for closer attention to the actual
language used by people have been made by nurses (Lanceley & Clark 2013; Peplau
1952/1988; Stein-Parbury 2014), psychiatrists (Lewis, Escalona & Keith 2009; Meares
2005), narrative therapists (Jenkins 1990; White 2007; White & Epston 1990),
researchers (Rochester & Martin 1979; Wengraf 2001) and linguists (Littlemore 2015;
Verschueren 2009). The mapping of metonymy’s influence from cognitive linguistics to
the experience of people who live with mental illness that is presented in this study can
contribute to this development.
The study supports the facilitation of people who live with mental illness engaging with
each other. This does not happen automatically, but requires support and facilitation,
as the participants in Barham and Hayward’s (1991) study confirmed. Participants in
the current study reported that engagement with peers was among the most helpful
factors in their recovery. They reported the pragmatic benefits of the knowledge
exchanged in these encounters, including how to manage the impact of psychosis by
strategically using metonymic language to describe and conceptually contain it.
Finally, increased community awareness of communicative competency across
different language abilities may contribute to expanding social networks, so that people
whose social intent is in fact inclusive are not unwittingly practising social exclusion
toward people who live with mental illness. Programs such as SANE’s StigmaWatch
Chapter 6 Discussion
171
(SANE Australia n.d.) can contribute, not simply through monitoring and educating
media, but by engaging the community in attending to the power of representations.
Conclusion
This study explored the influence of metonymy and pragmatic inferencing on the
experience of people who live with mental illness. The findings demonstrated that the
two research questions intersect, as the influence of metonymy on the speech
production and comprehension of people who live with mental illness has an impact on
the broader influence of metonymy in their experience, including how they give voice to
that experience.
The study collected data from people living with mental illness about their experience of
language in clinical and other settings, in their own words. It thus adds to knowledge,
beyond diagnostic or experimental results, information about people’s experience of
language in everyday life, and in their contact with the services designed to help them.
The study demonstrated that metonymy influences the everyday language of people
who live with mental illness, consistent with their membership of the broader speech
community. This contributes information about language use by a specific population to
the cognitive linguistic field.
The study further demonstrated that metonymic inferences carry specific significance in
the experience of people who live with mental illness. Inferences drawn about the
people who live with mental illness can inform clinical treatment including diagnosis,
and the subsequent delivery or withholding of care. Diagnoses in turn metonymically
influence the experiences of identity and stigma.
These inferences can be determined by metonymy operating conceptually or
linguistically, consciously or unconsciously, by people themselves, or by the people
they interact with, including clinicians. Each of these can be influenced by knowledge
and perceptions about the world, or meaning conveyed through the medium of
language.
In each instance of metonymy, options arise. Meaning can be conveyed literally instead
of indirectly. Inferences can be drawn that are not intended. Negative inferences can
be deliberately intended, but disavowed, creating uncertainty. Attributes can be
deemed salient and foregrounded, leading to efficiency on the one hand, but potential
mishaps on the other. It is neither possible, nor desirable, to reduce the influence of
Chapter 6 Discussion
172
metonymy. However, this study demonstrates that greater attendance to this influence
has the potential to improve the experience of people who live with mental illness.
The study demonstrated that participants themselves already express acute awareness
of the language they experience, and the influence of this on their own experience. In
response, they have developed sophisticated strategies in their own language use to
address these influences, contesting practices that hinder, and supporting and
reinforcing those that help.
Chapter 6 Discussion
173
Chapter 6 Discussion
174
APPENDICES
Appendix A: Information for participants
A Study of Metonymy in the Language of Mental Health Nursing
Information for Participants
Introduction
You are invited to take part in a research study into language use in communication
about mental health, because of your experience of living with a mental health problem,
and/or working with people with mental health problems.
Metonymy is the name for what we do when we call something by a word closely
related to the actual word, selecting an attribute of the thing to stand for the whole
thing. For example, if we say ‘give me a hand’ we are not literally asking for just one
body part, but for help from the whole person. Frequently, selected attributes are taken
to mean more about people than just the thing named. If we know a nurse is married to
an engineer, we are likely to think the nurse is female and the engineer male, based on
their occupations. Similarly, a person can be spoken of in terms of an illness they have:
‘an asthmatic’ or ‘a schizophrenic’.
This study is being conducted by Andrew Moors, a PhD student in the Faculty of
Nursing Midwifery and Health at the University of Technology, Sydney. He is being
supervised by Professor Jane Stein-Parbury, of the University of Technology, Sydney.
Study Procedures
If you agree to participate in this study, you will be asked to sign the Participant
Consent Form. You will then be asked to agree to participate in two interviews,
conducted approximately a month apart. In the first interview you will be asked to tell a
story about your experience, and in the second interview you will be asked questions
based on elements from your own story.
These interviews will be digitally audio recorded. These recordings will then be
transcribed word for word, with any information that may identify individuals removed.
The language used will then be analysed. The transcripts with no identifying material
will be stored in password-protected computers, and a locked filing cabinet, and only
the researcher and his supervisors will have access to them.
Risks
Appendices
175
Participation in this study should carry minimal risks for you. You may feel inhibited in
what you say because you are aware it is being recorded, or you may feel afterwards
that you wish you hadn’t said something in a certain way.
If you feel any discomfort, you can discuss this with Andrew Moors, or with your regular
treating clinicians.
You can ask to have your contributions erased from the record, before or during data
analysis, however once study results are published this would not be possible.
Benefits
While it is intended that this research study furthers knowledge about effective
communication about mental health issues, it may not be of direct benefit to you.
Costs
Participation in this study will not cost you anything, nor will you be paid.
Voluntary Participation
Participation in this study is entirely voluntary. You do not have to take part in it. If you
do take part, you can withdraw at any time without having to give a reason. Whatever
your decision, please be assured that it will not affect your treatment, or your
relationship with the staff who are caring for you.
Confidentiality
All the information collected from you for the study will be treated confidentially, and
only the researcher named above, and his supervisors, will have access to it. The
study results may be presented at a conference, in a thesis submitted for a Doctor of
Philosophy at the University of Technology, Sydney, or in a scientific publication, but
individual participants will not be identifiable in such a presentation.
Feedback to participants
A plain language summary of results will be available from the researcher at the
completion of the project which can be sent to you electronically via email or through
the post. If you would like to have a copy sent to you please provide Andrew Moors
with an email or postal address. These contact details will not be linked to the data and
will be destroyed once a summary has been sent out to you.
Further Information
When you have read this information, Andrew Moors will discuss it with you further,
and answer any questions you may have. If you would like to know more at any stage,
please feel free to contact him on 0400 84 22 66.
This information sheet is for you to keep.
Ethics Approval and Complaints
Appendices
176
This study has been approved by the Ethics Review Committee (RPAH Zone) of the
Sydney Local Health Network. Any person with concerns or complaints about the
conduct of this study should contact the Executive Officer on 02 9515 6766 and quote
protocol number X11-0031.
Appendices
177
Appendix B: Recruitment information published on websites
Language and Psychosis
What is the research project about?
The aim of this research is to gain an understanding of the influence of language on
people with lived experience of psychosis. In particular, I am looking at the language
that gets used about people in different contexts, including therapeutic settings, and if
this language leads to constraints on opportunities for recovery.
Who is doing the study?
My name is Andrew Moors. I am doing this research for a PhD in Nursing, supervised
by Professor Jane Stein-Parbury in the UTS Faculty of Health. I have worked as a
mental health nurse for seventeen years, and the research arises from my clinical
work.
What does the study involve?
The research entails participation in two interviews, held a month apart, and each
lasting between 30 to 60 minutes.
This research has been reviewed and approved by the Human Research Ethics
Committees of UTS and Sydney Local Health District.
Who am I looking for?
I am seeking participants with lived experience of psychosis, with or without a
diagnosis of schizophrenia, who would be comfortable telling their story, and then
answering some questions about their experience.
Why participate?
This research is aimed at better understanding specific aspects of communication
between people with lived experience of psychosis and others, including nurses,
narrative therapists and other health professionals. I hope to contribute to the
development of more effective therapeutic communications.
More information:
For further information about this study, please contact Andrew Moors at:
[email protected]
Appendices
178
Appendix C: Participant consent form
A Study of Metonymy in the Language of Mental Health Nursing
PARTICIPANT CONSENT FORM
I, ………………………………………………………………………………… [name]
of………………………………………………………………………………… [address]
Have read and understood the Information for Participants on the above named
research study and have discussed the study with …………………………………
I have been made aware of the procedures involved in the study, including any known
or expected inconvenience, risk, discomfort or potential side effect and of their
implications as far as they are currently known by the researchers.
I understand that the interviews will be digitally audio recorded, and the recordings
electronically archived, and I agree to this.
I freely choose to participate in this study and understand that I can withdraw at any
time.
I also understand that the research study is strictly confidential.
I hereby agree to participate in this research study.
NAME:
…………………………………………………………………..
SIGNATURE:
…………………………………………………………………..
DATE:
…………………………………………………………………..
NAME OF WITNESS:
SIGNATURE OF WITNESS:
Appendices
…………………………………………………………..
…………………………………………………..
179
Appendices
180
REFERENCES
Ahlsen, E. 2006, Introduction to Neurolinguistics, John Benjamins Publishing
Company, Amsterdam, NLD.
Akiskal, H.S. 2009, 'Mood Disorders: Clinical Features', in B.J. Sadock, V.A.
Sadock & P. Ruiz (eds), Kaplan & Sadock's Comprehensive Textbook of
Psychiatry, vol. 1, Lippincott WIlliams and Wilkins, Philadelphia, pp.
1693-733.
Akmajian, A., Demers, R.A., Farmer, A.K. & Harnish, R.M. 2010, Linguistics: An
Introduction to Language and Communication, 6th edn, The MIT Press,
Cambridge, MA.
Al-Sharafi, A.G.M. 2004, Textual metonymy : a semiotic approach, Palgrave
Macmillan, Houndmills, Basingstoke, Hampshire ; New York.
Alford, D.K.H. 1978, 'The Demise of the Whorf Hypothesis (A Major Revision in
the History of Linguistics)', paper presented to the 4th Annual Meeting of
the Berkely Linguistics Society, Berkeley.
American Psychiatric Association 2013, Diagnostic and statistical manual of
mental disorders: DSM-5, 5 edn, American Psychiatric Association,
Arlington.
Andreasen, N.C. 1979a, 'Thought, Language and Communication Disorders I.
Clinical Assessment, Definition of Terms, and Evaluation of Their
Reliability', Archives of General Psychiatry, vol. 36, pp. 1315-21.
Andreasen, N.C. 1979b, 'Thought, Language and Communication Disorders II.
Diagnostic Significance', Archives of General Psychiatry, vol. 36, pp.
1325-30.
Andreasen, N.C. & Grove, W.M. 1986, 'Thought, Language, and
Communication in Schizophrenia: Diagnosis and Prognosis',
Schizophrenia Bulletin, vol. 12, no. 3, pp. 348-59.
Arntfield, M. 2008, 'Hegemonic Shorthand: Technology and Metonymy in
Modern Policing', The Communication Review, vol. 11, no. 1, pp. 76-97.
Australian Commission on Safety and Quality in Health Care 2014, Scoping
Study on the Implementation of National Standards in Mental Health
Services, ACSQHC, Sydney.
Barcelona, A. 2011, 'Reviewing the properties and prototype structure of
metonymy', in R. Benczes, A. Barcelona & F.J. Ruiz de Menoza Ibanez
(eds), Defining Metonymy in Cognitive Linguistics: Towards a consensus
view, John Benjamins, Amsterdam, pp. 7-57.
Barham, P. & Hayward, R. 1991, From the Mental Patient to the Person,
Routledge, London.
Barker, P. 2001, 'The Tidal Model: developing an empowering, person-centred
approach to recovery within psychiatric and mental health nursing',
Journal of Psychiatric and Mental Health Nursing, vol. 8, pp. 233-40.
Barnes, M. & Berke, J. 1971, Two accounts of a journey through madness,
Harcourt Brace Jovanovich, Oxford.
Barthes, R. 1994, The Semiotic Challenge, trans. R. Howard, University of
California Press, Berkeley.
References
181
Becker, H.S. 1963, Outsiders: Studies in the Sociology of Deviance, The Free
Press, New York.
Benczes, R., Barcelona, A. & Ruiz de Menoza Ibanez, F.J. (eds) 2011, Defining
Metonymy in Cognitive Linguistics: Towards a consensus view, John
Benjamins, Amsterdam.
Bird, J. 2000, The heart's narrative : therapy and navigating life's contradictions,
Edge Press, Auckland, N.Z.
Bleuler, E. 1911/1950, Dementia Praecox or the Group of Schizophrenias,
International Universities Press, New York.
Bowie, C.J. & Harvey, P.D. 2008, 'Communication abnormalities predict
functional outcomes in chronic schizophrenia: Differential associations
with social and adaptive functions', Schizophrenia Research, vol. 103,
pp. 240-7.
Brdar-Szabo, R. & Brdar, M. 2011, 'What do metonymic chains reveal about the
nature of metonymy?', in R. Benczes, A. Barcelona & F.J. Ruiz de
Menoza Ibanez (eds), Defining Metonymy in Cognitive Linguistics:
Towards a Consensus View, John Benjamins, Amsterdam, pp. 217-48.
Brdar-Szabo, R. & Brdar, M. 2012, 'The problem of data in the cognitive
linguistic research on metonymy: a cross-linguistic perspective',
Language Sciences, vol. 34, pp. 728-45.
Brdar, M. & Brdar-Szabo, R. 2003, 'Metonymic coding of linguistic action in
English, Croatian and Hungarian', in K.-U. Panther & L.L. Thornburg
(eds), Metonymy and Pragmatic Inferencing, John Benjamins,
Amsterdam, pp. 241-66.
Brown, W. & Kandirikirira, N. 2007, Recovering mental health in Scotland.
Report on narrative investigation of mental health recovery, Glasgow.
Brune, M. 2005, '"Theory of Mind" in Schizophrenia: A Review of the Literature',
Schizophrenia Bulletin, vol. 31, no. 1, pp. 21-42.
Bruner, J. 1997, 'Labov and Waletzky Thirty Years On', Journal of Narrative and
Life History, vol. 7, no. 1-4, pp. 61-8.
Bruner, J.S. 1990, Acts of meaning, Harvard University Press, Cambridge,
Mass.
Burns, E., Fenwick, J., Sheehan, A. & Schmied, V. 2012, 'Mining for liquid gold:
midwifery language and practices associated with early breastfeeding
support', Maternal and Child Nutrition, vol. 9, pp. 57-73.
Bury, M. 2001, 'Illness narratives: fact or fiction?', Sociology of Health and
Illness, vol. 23, no. 3, pp. 263-85.
Butler, J. 1990, Gender trouble and the subversion of identity, Routlege, New
York & London.
Butler, J. 1993, Bodies that Matter: On the discursive limits of "sex", Routledge,
New York.
Cameron, N. 1938, 'Reasoning, regression and communication in
schizophrenics', Psychological Monographs, vol. 50, no. 1, pp. i-34.
Cameron, N. 1944, 'Experimental Analysis of Schizophrenic Thinking', in J.S.
Kasanin (ed.), Language and Thought in Schizophrenia, University of
California Press, Berkeley, pp. 50-64.
Candlin, S. 2008, Therapeutic communication : a lifespan approach, Pearson
Education Australia, Frenchs Forest, N.S.W.
References
182
Cashin, A., Browne, G., Bradbury, J. & Mulder, A. 2013, 'The Effectiveness of
Narrative Therapy With Young People With Autism', Journal of Child and
Adolescent Psychiatric Nursing, vol. 26, pp. 32-41.
Cavell, S. 1958, 'Must we mean what we say?', Inquiry: An Interdisciplinary
Journal of Philosophy, vol. 1, no. 1-4, pp. 172-212.
Champagne-Lavau, M. & Stip, E. 2010, 'Pragmatic and executive dysfunction in
schizophrenia', Journal of Neurolinguistics, vol. 23, pp. 285-96.
Chapman, L.J. 1960, 'Confustion of Figurative and Literal Usages of Words by
Schizophrenics and Brain Damaged Patients', Journal of Abnormal and
Social Psychology, vol. 60, no. 3, pp. 412-6.
Chomsky, N. 1988, Language and Problems of Knowledge: The Managua
Lectures, The MIT Press, Cambridge.
Cienki, A. 2007, 'Frames, Idealized Cognitive Models, and Domains', in D.
Geeraerts & H. Cuyckens (eds), Oxford Handbook of Cognitive
Linguistics, Oxford University Press, Oxford, pp. 170-87.
Cleary, M., Hunt, G.E., Escott, P. & Walter, G. 2010, 'Receiving difficult news.
Views of patients in an inpatient setting', Journal of psychosocial nursing
and mental health services, vol. 48, no. 6, pp. 40-8.
Clegg, J., Brumfitt, S., Parks, R.W. & Woodruff, P.W.R. 2007, 'Speech and
language therapy intervention in schizophrenia: a case study',
International Journal of Language and Communication Disorders, vol. 42,
no. S1, pp. 81-101.
Corley, M. 2002, 'Nurse Moral Distress: A Proposed Theory and Research
Agenda', Nursing Ethics, vol. 9, no. 6, pp. 636-50.
Corrigan, P.W., Kosyluk, K.A. & Rusch, N. 2013, 'Reducing Self-Stigma by
Coming Out Proud', American Journal of Public Health, March 14 2013,
pp. e1-e7.
Costa, L., Voronka, J., Landry, D., Reid, J., McFarlane, B., Reville, D. & Church,
K. 2012, 'Recovering our Stories: A Small Act of Resistance', Studies in
Social Justice, vol. 6, no. 1, pp. 85-101.
Crawford, P., Johnson, A.J., Brown, B.J. & Nolan, P. 1999, 'The language of
mental health nursing reports: firing paper bullets?', Journal of Advanced
Nursing, vol. 29, no. 2, pp. 331-140.
Crawford, R.M. 2014, 'Emotional communication between nurses and parents of
a child in hospital', PhD thesis, University of Technology, Sydney.
Croft, W. 2006, 'On explaining metonymy: Comment on Peirsman and
Geeraerts, "Metonymy as a prototypical category"', Cognitive Linguistics,
vol. 17, no. 3, pp. 317-26.
Croft, W. 2009, 'Toward a social cognitive linguistics', in V. Evans & S. Pourcel
(eds), New Directions in Cognitive Linguistics, John Benjamins,
Amsterdam, pp. 395-420.
Croft, W. & Cruse, D.A. 2004, Cognitive Linguistics, Cambridge University
press, Cambridge.
Crow, T.J. 2010, 'The nuclear symptoms of schizophrenia reveal the four
quadrant structure of language and its deictic frame', Journal of
Neurolinguistics, vol. 23, pp. 1-9.
De Man, P. 1979, Allegories of reading : figural language in Rousseau,
Nietzsche, Rilke, and Proust, Yale University Press, New Haven.
Dellar, R., Curtis, T. & Leslie, E. (eds) 2000, Mad Pride: A Celebration of Mad
Culture, Chipmunka Publishing.
References
183
Deveson, A. 2003, Resilience, Allen & Unwin, Crows Nest.
Docherty, N.M. 2012, 'On Identifying the Processes Underlying Schizophrenic
Speech Disorder', Schizophrenia Bulletin, vol. 38, no. 6, pp. 1327-35.
Eggins, S. & Slade, D. 1997, Analysing casual conversation, Cassell, London ;
New York.
Epstein, M. n.d., A Consumer Activist's Guide to Mental Health in Australia,
weblog, viewed 19 April 2015, <http://www.takver.com/epstein/>.
Evans, V. 2007, A Glossary of Cognitive Linguistics, Edinburgh University
Press, Edinburgh.
Fauconnier, G. 1997, Mappings in thought and language, Cambridge University
Press, Cambridge, U.K. ; New York, NY, USA.
Feyaerts, K. 2000, 'Refining the Inheritance Hypothesis: Interaction between
metaphoric and metonymic hierarchies', in A. Barcelona (ed.), Metaphor
and Metonymy at the Crossroads, Mouton de Gruyter, Berlin, pp. 59-78.
Fine, J. 2006, Language in Psychiatry: A Handbook of Clinical Practice,
Equinox, London.
Fischer-Rosenthal, W. 2000, 'Biographical work and biographical structuring in
present-day societies', in P. Chamberlayne, J. Bornat & T. Wengraf
(eds), The turn to biographical methods in social science: comparitive
issues today, Routledge, Abingdon, pp. 109-25.
Flaskas, C. 2002, Family therapy beyond postmodernism : practice challenges
theory, Brunner-Routledge, Hove, East Sussex.
Foucault, M. 1980, Power/Knowledge: Selected interviews and other writings,
1972-1977, Harvester Press, Brighton.
Freshwater, D. & Stickley, T. 2003, 'The heart of the art: emotional intelligence
in nurse education', Nursing Inquiry, vol. 11, no. 2, pp. 91-8.
Fusar-Poli, P. & Politi, P. 2008, 'Paul Eugen Bleuler and the Birth of
Schizophrenia (1908)', American Journal of Psychiatry, vol. 165, p. 1407.
Geeraerts, D. 2008, 'Introduction: A rough guide to Cognitive Linguistics', in D.
Geeraerts (ed.), Cognitive Linguistics: Basic Readings, De Gruyter
Mouton, Berlin, pp. 1-29.
Geeraerts, D. & Cuyckens, H. 2007, 'Introducing Cognitive Linguistics', in D.
Geeraerts & H. Cuyckens (eds), The Oxford Handbook of Cognitive
Linguistics, Oxford University Press, Oxford, pp. 3-21.
Geeraerts, D. & Peirsman, Y. 2011, 'Zones, facets, and prototype-based
metonymy', in R. Benczes, A. Barcelona & F.J. Ruiz de Menoza Ibanez
(eds), Defining Metonymy in Cognitive Linguistics: Towards a consensus
view, John Benjamins, Amsterdam, pp. 89-102.
Geertz, C. 1973, The Interpretation of Cultures, Basic Books, New York.
Gibbs, R. 2007a, 'Idioms and Formulaic Language', in D. Geeraerts & H.
Cuyckens (eds), The Oxford Handbook of Cognitive Linguistics, Oxford
University Press, Oxford, pp. 697-725.
Gibbs, R. 2007b, 'Why cognitive linguists should care more about empirical
methods', in M. Gonzalez-Marquez, I. Mittelberg, S. Coulson & M.J.
Spivey (eds), Methods in Cognitive Linguistics, John Benjamins,
Amsterdam, pp. 2-18.
Gibbs, R.W. 2007c, 'Experimental tests of figurative meaning construction', in
G. Radden (ed.), Aspects of meaning construction, John Benjamins,
Amsterdam, pp. 19-32.
References
184
Gibbs, R.W. & Colston, H.L. 2006, 'Figurative Language', in M.J. Traxler & M.A.
Gernsbacher (eds), Handbook of Psycholinguistics, 2nd edn, Academic
Press, London, pp. 835-62.
Giora, R. 2003, On our mind : salience, context, and figurative language, Oxford
University Press, Oxford ; New York.
Goffman, E. 1963, Stigma: Notes on the Management of Spoiled Identity,
Prentice Hall, Englewood Cliffs.
Goffman, E. 1968, Asylums: Essays on the social situation of mental patients
and other inmates, Penguin Books, Harmondsworth.
Goldstein, K. 1944, 'Methodological approach to the study of schizophrenic
thought disorder', in J.S. Kasanin (ed.), Language and thought in
schizophrenia, University of California Press, Berkeley, pp. 17-40.
Green, M.F. & Harvey, P.D. 2014, 'Cognition in schizophrenia: Past, present,
and future', Schizophrenia Research: Cognition, vol. 1, pp. 1-9.
Grice, H.P. 1975, 'Logic and conversation', in P. Cole & J.L. Morgan (eds),
Syntax and Semantics. Vol.3: Speech Acts, vol. 3, Academic Press, New
York, pp. 41-58.
Halliday, M.A.K. & Matthiessen, C.M.I.M. 2004, An introduction to functional
grammar, 3rd edn, Arnold, London.
Hamilton, B. & Manias, E. 2006, ''She's manipulative and he's right off': A critical
analysis of psychiatric nurses' oral and written language in the acute
inpatient setting', International Journal of Mental Health Nursing, vol. 15,
pp. 84-92.
Hammersley, M. 2008, Questioning Qualitative Inquiry, Sage, London.
Hanisch, C. 2006, 'The Personal is Political', 2006, viewed 25 March 2015,
<http://www.carolhanisch.org/CHwritings/PersonalisPol.pdf>.
Hardie, M.J. 1999, 'Beard', in J. Selzer & S. Crowley (eds), Rhetorical Bodies,
University of Wisconsin Press, Madison, pp. 275-96.
Harris, R.A. 1993, The Linguistics Wars, Oxford University Press, Oxford.
Henderson-Brooks, C. 2010, ''Words being its marker': a linguistic study of self
as shifting state in three types of therapeutic conversation', in E. Swain
(ed.), Thresholds and Potentialities of Systemic Functional Linguistics:
Multilingual, Multimodal and Other Specialised Discourses, Edizioni
Universita di Trieste, Trieste, pp. 229-67.
Hill, T.E. 2010, 'How clinicians make (or avoid) moral judgements of patients:
implications of the evidence for relationships and research', Philosophy,
Ethics, and Humanities in Medicine, vol. 5, no. 11, p. 14.
Hobbes, T. 1650/2011, Leviathan, Broadview Press, Peterborough.
Holshausen, K. 2012, 'Structured Therapy for the Enhancement of Purposeful
Speech (STEPS): A Step in the Right Direction to Treating Formal
Thought Disorder in Schizophrenia', M. Psychology thesis, Queen's
University, Kingston, Ontario.
Hudson, R. 2007, 'Word Grammar', in D. Geeraerts & H. Cuyckens (eds),
Oxford Handbook of Cognitive Linguistics, Oxford University Press,
Oxford, pp. 509-40.
Hunter, L.P. 2006, 'Women give birth and pizzas are delivered: Language and
Western childbirth paradigms', Journal of Midwifery and Women's Health,
vol. 51, no. 2, pp. 119-24.
References
185
Jakobson, R. 1956/1987, 'Two Aspects of Language and Two Types of Aphasic
Disturbances', in R. Jakobson, K. Pomorska & S. Rudy (eds), Language
in Literature, Belknap Press, Cambridge, Mass., pp. 95-114.
Jakobson, R., Pomorska, K. & Rudy, S. 1987, Language in literature, Belknap
Press, Cambridge, Mass.
James, W. 1890/1950, The principles of psychology, vol. 1, Dover Pubs., New
York.
James, W. 1907/1987, 'Pragmatism', in B. Kuklick (ed.), Writings 1902-1910,
Library of America, New York, pp. 479-625.
Jenkins, A. 1990, Invitations to responsibility : the therapeutic engagement of
men who are violent and abusive, Dulwich Centre Publications, Adelaide.
King, N. 2004, 'Using templates in the thematic analysis of text', in C. Cassell &
G. Symon (eds), Essential Guide To Qualitative Methods in
Organizational Research, Sage, London, pp. 256-70.
Kingdon, D.G., Kinoshita, Y., Naeem, F., Swelan, M. & Hansen, L. 2007,
'Schizophrenia can and should be renamed', British Medical Journal, vol.
334, pp. 221-2.
Kovecses, Z. 2006, Language, mind and culture: a practical introduction, Oxford
University Press, New York.
Kraepelin, E. 1919/1999, Dementia praecox and paraphrenia, trans. R.M.
Barclay, Kreiger Publishing, Huntington.
Kuperberg, G. 2010a, 'Language in Schizophrenia Part 1: An Introduction',
Language and Linguistics Compass, vol. 4, no. 8, pp. 576-89.
Kuperberg, G. 2010b, 'Language in Schizophrenia Part 2: What Can
Psycholinguistics Bring to the Study of Schizophrenia...and Vice Versa?',
Language and Linguistics Compass, vol. 4, no. 8, pp. 590-604.
Kuperberg, G. & Caplan, D. 2003, 'Language Dysfunction in Schizophrenia', in
R.B. Schiffer, S.M. Rao & B.S. Fogel (eds), Neuropsychiatry, 2nd edn,
Lippincott Williams and Wilkins, Philadelphia, pp. 444-66.
Kuperberg, G., Kreher, D.A. & Ditman, T. 2009, 'What can Event-related
Potentials tell us about language, and perhaps even thought, in
schizophrenia?', International Journal of Psychophysiology, vol. 75, no.
2, pp. 66-76.
Labov, W. & Waletzky, J. 1967/1997, 'Narrative Analysis: Oral Version of
Personal Experience', Journal of Narrative and Life History, vol. 7, no. 14, pp. 3-38.
Lacan, J. 2006, Ecrits: The First Complete Edition in English, trans. B. Fink, W.
W. Norton, New York.
Laing, R.D. 1960, The Divided Self: An Existential Study in Sanity and
Madness, Penguin, Harmondsworth.
Lakoff, G. 1987, Women, fire, and dangerous things : what categories reveal
about the mind, University of Chicago Press, Chicago.
Lakoff, G. & Johnson, M. 1980, Metaphors we live by, University of Chicago
Press, Chicago.
Lanceley, A. & Clark, J.M. 2013, 'Cancer in other words? the role of metaphor in
emotion disclosure in cancer patients', British Journal of Psychotherapy,
vol. 29, no. 2, pp. 182-201.
Langacker, R.W. 1993, 'Reference-point constructions', Cognitive Linguistics,
vol. 4, no. 1, pp. 1-38.
References
186
Langacker, R.W. 2007, 'Cognitive Grammar', in D. Geeraerts & H. Cuyckens
(eds), Oxford Handbook of Cognitive Linguistics, Oxford University
Press, Oxford, pp. 421-62.
Langacker, R.W. 2009, Investigations in Cognitve Grammar, Mouton de
Gruyter, Berlin.
Larsen, N. 1929, Passing, Knopf, New York.
Lewis, S., Escalona, P.R. & Keith, S.J. 2009, 'Phenomenology of
Schizophrenia', in B.J. Sadock, V.A. Sadock & P. Ruiz (eds), Kaplan &
Sadock's Comprehensive Textbook of Psychiatry, 9th edn, vol. 1,
Lippincott Williams and Wilkins, Philadelphia, pp. 1434-51.
Littlemore, J. 2015, Metonymy: Hidden Shortcuts in Language, Thought and
Communication, Cambridge University Press, Cambridge.
Lloyd-Evans, B., Mayo-Wilson, E., Harrison, B., Istead, H., Brown, E., Pilling, S.,
Johnson, S. & Kendall, T. 2014, 'A systematic review and meta-analysis
of randomised controlled trials of peer support for people with severe
mental illness', BMC psychiatry, vol. 14, no. 1, p. 39.
Lodge, D. 1992, The Art of Fiction, Penguin, Harmondsworth.
Lysaker, P.H., Carcione, A., Dimaggio, G., Johannesen, J.K., Nicolo, G.,
Procacci, M. & Semerari, A. 2005, 'Metacognition amidst narratives of
self and illness in schizophrenia: associations with neurocognition,
symptoms, insight and quality of life', Acta Psychiatrica Scandinavica,
vol. 112, pp. 64-71.
Lysaker, P.H., Roe, D. & Yanos, P.T. 2006, 'Toward Understanding the Insight
Paradox: Internalized Stigma Moderates the Association Between Insight
and Social Functioning, Hope, and Self-Esteem Among People with
Schizophrenia Spectrum Disorders', Schizophrenia Bulletin, vol. 33, no.
1, pp. 192-9.
Markert, K. & Nissim, M. 2003, 'Corpus-Based Metonymy Analysis', Metaphor
and Symbol, vol. 18, no. 3, pp. 175-88.
McGurk, S., Twamley, E.W., Sitzer, D.I., McHugo, G.J. & Mueser, K.T. 2007, 'A
Meta-Analysis of Cognitive Remediation in Schizophrenia', American
Journal of Psychiatry, vol. 164, no. 12, pp. 1791-802.
McKenna, P. & Oh, T. 2005, Schizophrenic Speech: Making Sense of Bathroots
and Ponds that Fall in Doorways, Cambridge University Press,
Cambridge.
Meagher, J. 2014, 'Thriving not just surviving', World Mental Health Day 2014:
Living with Schizophrenia, World Federation for Mental Health,
Occoquan, VA, pp. 8-9.
Meares, R. 2005, The metaphor of play : origin and breakdown of personal
being, 3rd edn, Routledge, New York.
Meares, R., Butt, D., Henderson-Brooks, C. & Samir, H. 2005, 'A poetics of
change', Psychoanalytic Dialogues, vol. 15, no. 5, pp. 661-80.
Melogno, S. & Pinto, M.A. 2014, 'Enhancing Metaphor and Metonymy
Comprehension in Children with High-Functioning Autism Spectrum
Disorder', Psychology, vol. 5, pp. 1375-83.
Mendoza, J., Bresnan, A., Rosenberg, S., Elson, A., Gilbert, Y., Long, P.,
Wilson, K. & Hopkins, J. 2013, Obsessive Hope Disorder: Reflections on
30 years of mental health reform in Australia and visions for the future.
Summary Report, Caloundra.
Mental Health Act (Victoria) 2014.
References
187
Mental Health Act (Western Australia) 2014.
Mental Health and Drug and Alcohol Office 2008, Information Bulletin:
Redesigned Mental Health Clinical Documentation: Notification of
Availability, IB2008_047, Department of Health, NSW, North Sydney,
<www0.health.nsw.gov.au/policies/ib/2008/pdf/IB2008_047.pdf>.
Menzies Lyth, I. 1959/1988, 'The functioning of social systems as a defence
against anxiety', in I. Menzies Lyth (ed.), Containing Anxiety in
Institutions, vol. 1, Free Association Books, London, pp. 43-85.
Menzies Lyth, I. 1989, The Dynamics of the Social: Selected Essays, vol. 2,
Free Association Books, London.
Milton, A.C. & Mullan, B.A. 2014, 'A Qualitative Exploration of Service Users’
Information Needs and Preferences When Receiving a Serious Mental
Health Diagnosis', Community mental health journal, vol. 51, no. 4, pp.
459-66.
Minas, H., Kakuma, R., Too, L.S., Vayani, H., Orapeleng, S., Prasad-Ildes, R.,
Turner, G., Procter, N. & Oehm, D. 2013, 'Mental health research and
evaluation in multicultural Australia: developing a culture of inclusion',
International Journal of Mental Health Systems, vol. 7, no. 23, pp. 1-25,
<http://www.biomedcentral.com/content/pdf/1752-4458-7-23.pdf>.
Minichiello, V., Madison, J., Hays, T. & Parmenter, G. 2004, 'Doing qualitative
in-depth interviews', in V. Minichiello, G. Sullivan, K. Greenwood & R.
Axford (eds), Handbook of research methods for nursing and health
science, 2nd edn, Pearson Education Australia, Frenchs Forest, pp. 41146.
Mueser, K.T., Pratt, S.I., Bartels, S.J., Forester, B., Wolfe, R. & Cather, C. 2010,
'Neurocognition and social skill in older persons with schizophrenia and
major mood disorders: An analysis of gender and diagnosis effects',
Journal of Neurolinguistics, vol. 23, pp. 297-317.
Muntigl, P. 2004, Narrative Counselling: Social and linguistic processes of
change, John Benjamins, Amsterdam.
National Health and Medical Research Council, Australian Research Council &
Australian Vice-Chancellor's Committee 2007/2015, National Statement
on Ethical Conduct in Human Research 2007 (Updated 2015), E72,
Commonwealth of Australia, Canberra.
National Mental Health Commission 2012, A Contributing Life, the 2012
National Report Card on Mental Health and Suicide Prevention, NMHC,
Sydney.
National Mental Health Commission 2014, The National Review of Mental
Health Programmes and Services, vol. 3, National Mental Health
Commission, Sydney.
Nunberg, G. 1995, 'Transfers of Meaning', Journal of Semantics, vol. 12, pp.
109-32.
Oxford Dictionaries 2015, Oxford University Press, viewed 20 January 2015,
<Error! Hyperlink reference not valid..
Panther, K.-U. & Thornburg, L.L. 2003a, 'Introduction: On the nature of
conceptual metonymy', in K.-U. Panther & L.L. Thornburg (eds),
Metonymy and Pragmatic Inferencing, John Benjamins, Amsterdam, pp.
1-20.
Panther, K.-U. & Thornburg, L.L. (eds) 2003b, Metonymy and Pragmatic
Inferencing, John Benjamins, Amsterdam.
References
188
Panther, K.-U. & Thornburg, L.L. 2007, 'Metonymy', in D. Geeraerts & H.
Cuyckens (eds), Oxford Handbook of Cognitive Linguistics, Oxford
University Press, Oxford, pp. 236-63.
Patel, V.L., Kannampallil, T.G. & Shortliffe, E.H. 2015, 'Role of cognition in
generating and mitigating clinical errors', BMJ Quality and Safety, 2 May
2015.
Peirsman, Y. & Geeraerts, D. 2006a, 'Don't let metonymy be misunderstood: an
answer to Croft', Cognitive Linguistics, vol. 17, no. 3, pp. 327-35.
Peirsman, Y. & Geeraerts, D. 2006b, 'Metonymy as a protoypical category',
Cognitive Linguistics, vol. 17, no. 3, pp. 269-316.
Peplau, H.E. 1952/1988, Interpersonal Relations in Nursing: A Conceptual
Frame of Reference for Psychodynamic Nursing, Macmillan, London.
Preminger, A., Warnke, F.J. & Hardison, O.B. 1986, The Princeton handbook of
poetic terms, Princeton University Press, Princeton, N.J.
QSR International 2012, 'NVivo qualitative data analysis software, Version 10',
QSR International Pty Ltd.
Radden, G. & Kovecses, Z. 1999, 'Towards a Theory of Metonymy', in K.-U.
Panther & G. Radden (eds), Metonymy in Language and Thought, John
Benjamins Publishing Company, Amsterdam, pp. 17-59.
Radden, G. & Seto, K.-i. 2003, 'Metonymic construals of shopping requests in
have- and be- languages', in K.-U. Panther & L.L. Thornburg (eds),
Metonymy and Pragmatic Inferencing, John Benjamins, Amsterdam, pp.
223-39.
Rapp, A., Hensler, M., Bartels, M., Mutschler, D., Saur, R. & Markert, K. 2008,
'Metonymy resolution in schizophrenia: A model for complex semantic
language comprehension', Schizophrenia Research, vol. 102, no. 1–3,
Supplement 2, p. 144.
Rapp, A.M., Erb, M., Grodd, W., Bartels, M. & Markert, K. 2011, 'Neural
correlates of metonymy resolution', Brain and Language, vol. 119, no. 3,
pp. 196-205.
Reavley, N.J. & Jorm, A.F. 2011, National Survey of Mental Health Literacy and
Stigma, Department of Health and Ageing, Canberra.
Repper, J. & Carter, T. 2011, 'A review of the literature on peer support in
mental health services', Journal of Mental Health, vol. 20, no. 4, pp. 392411.
Rhodes, J.E. & Jakes, S. 2004, 'The contribution of metaphor and metonymy to
delusions', Psychology and Psychotherapy: Theory, Research and
Practice, vol. 77, pp. 1-17.
Richards, I.A. 1991, Richards on Rhetoric: I A Richards, Selected Essays 19291974, Oxford University Press, Oxford.
Ricoeur, P. 1975/2003, The Rule of Metaphor: The creation of meaning in
language, trans. R. Czerny, K. McLaughlin & J. Costello, Routledge,
Abingdon.
Rochester, S. & Martin, J.R. 1979, Crazy talk: a study of the discourse of
schizophrenic speakers, Plenum Press, New York.
Rosch, E. 1978, 'Principles of Categorization', in E.R.a.B. Lloyd (ed.), Cognition
and Categorization, Lawrence Erlbaum Associates, Hillsdale New
Jersey, pp. 28-48.
References
189
Rosen, A., Hadzi-Pavlovic, D. & Parker, G. 1989, 'The Life Skills Profile: A
Measure Assessing Function and Disability in Schizophrenia',
Schizophrenia Bulletin, vol. 15, no. 2, pp. 325-37.
Ross, C.A. & Goldner, E.M. 2009, 'Stigma, negative attitudes and discrimination
towards mental illness within the nursing profession: a review of the
literature', Journal of Psychiatric and Mental Health Nursing, vol. 16, pp.
558-67.
Rüsch, N., Abbruzzese, E., Hagedorn, E., Hartenhauer, D., Kaufmann, I.,
Curschellas, J., Ventling, S., Zuaboni, G., Bridler, R. & Olschewski, M.
2014, 'Efficacy of Coming Out Proud to reduce stigma’s impact among
people with mental illness: pilot randomised controlled trial', The British
Journal of Psychiatry, vol. 204, no. 5, pp. 391-7.
Sadock, B.J. 2009, 'Signs and symptoms in psychiatry', in B.J. Sadock, V.A.
Sadock & P. Ruiz (eds), Kaplan & Sadock's Comprehensive Textbook of
Psychiatry, 9th edn, vol. 1, Lippincott Williams and Wilkins, Philadelphia,
pp. 918-29.
SANE Australia n.d., Stigma Watch, viewed 28 May 2015 2015,
<www.sane.org/stigmawatch>.
Saussure de, F. 1916/1974, Course in General Linguistics, trans. W. Baskin,
Fontana/Collins, London.
Scheff, T.J. 1974, 'The labelling theory of mental illness', American Sociological
Review, vol. 39, no. June, pp. 444-52.
Schmid, H.-J. 2012, 'Generalizing the apparently ungeneralizable. Basic
ingredients of a cognitive-pragmatic approach to the construal of
meaning-in-context', in H.-J. Schmid (ed.), Cognitive Pragmatics, De
Gruuter Mouton, Berlin, pp. 3-22.
Shefer, G., Henderson, C., Howard, L.M., Murray, J. & Thornicroft, G. 2014,
'Diagnostic Overshadowing and Other Challenges Involved in the
Diagnostic Process of Patients with Mental Illness Who Present in
Emergency Departments with Physical Symptoms - A Qualitative Study',
PLoS ONE, vol. 9, no. 11, pp. 1-9.
Slade, M. 2009, Personal recovery and mental illness: a guide for mental health
professionals, Cambridge University Press, Cambridge.
Speedy, J. 2005, 'A day in the life: a worm's eye view of doing collaborative
counselling research', Counselling & Psychotherapy Research, vol. 5, no.
1, pp. 75-6.
Squire, C. 2000, 'Situated selves, the coming-out genre and equivalent
citizenship in narratives of HIV', in P. Chamberlayne, J. Bornat & T.
Wengraf (eds), The Turn to Biographical Methods in Social Science:
Comparative Issues and Examples, Routledge, Abingdon, pp. 196-213.
Stein-Parbury, J. 2014, Patient and person: interpersonal skills in nursing, 5th
edn, Elsevier, Chatswood.
Stuhlmiller, C.M. 2001, 'Narrative methods in qualitative research: potential for
therapeutic transformation', in K.R. Gilbert (ed.), The Emotional Nature of
Qualitative Research, CRC Press, Boca Raton, pp. 63-80.
Takahashi, H., Ideno, T., Okubo, S., Matsui, H., Takemura, K., Matsuura, M.,
Kato, M. & Okubo, Y. 2009, 'Impact of changing the Japanese term for
"schizophrenia" for reasons of stereotypical beliefs of schizophrenia in
Japanese youth', Schizophrenia Research, vol. 112, pp. 149-52.
References
190
Talmy, L. 2007, 'Attention Phenomena', in D. Geeraerts & H. Cuyckens (eds),
Oxford Handbook of Cognitive Linguistics, Oxford University Press,
Oxford, pp. 264-93.
Tan, E., Thomas, N. & Rossell, S.L. 2014, 'Speech disturbances and quality of
life in schizophrenia: Differential impacts on functioning and life
satisfaction', Comprehensive Psychiatry, vol. 55, pp. 693-8.
Thalmann, W.G. 1984, Conventions of form and thought in early Greek epic
poetry, Johns Hopkins University Press, Baltimore.
Thornicroft, G., Rose, D., Kassam, A. & Sartorius, N. 2007, 'Stigma: ignorance,
prejudice or discrimination?', British Journal of Psychiatry, vol. 190, pp.
192-3.
Tranulis, C., Lecomte, T., El-Khoury, B., Lavarenne, A. & Brodeur-Cote, D.
2013, 'Changing the Name of Schizophrenia: Patient Perspectives and
Implications for DSM-V', PLoS ONE, vol. 8, no. 2, p. e55998.
Turner, M. 2007, 'Conceptual Integration', in D. Geeraerts & H. Cuyckens (eds),
Oxford Handbook of Cognitive Linguistics, Oxford University Press,
Oxford, pp. 377-93.
Uitterhoeve, R., Bensing, J., Dilven, E., Donders, R., deMulder, P. & van
Achterberg, T. 2009, 'Nurse–patient communication in cancer care: does
responding to patient's cues predict patient satisfaction with
communication', PsychoǦOncology, vol. 18, no. 10, pp. 1060-8.
Veale, T., Feyaerts, K. & Brone, G. 2006, 'The cognitive mechanisms of
adversarial humor', Humor, vol. 19, no. 3, pp. 305-39.
Verschueren, J. 2009, 'Introduction: The pragmatic perspective', in J.
Verschueren & J.-O. Ostman (eds), Key Notions for Pragmatics, John
Benjamins, Amsterdam, pp. 1-27.
Vygotsky, L. 1934/1994, 'Thought in Schizophrenia', in R. van der Veer & J.
Valsiner (eds), The Vygotsky Reader, Blackwell, Oxford, pp. 313-26.
Vygotsky, L. 1978, Mind in Society, Harvard University Press, Cambridge MA.
Vygotsky, L.S. & Kozulin, A. 1934/1986, Thought and language, Translation
newly rev. and edited / edn, MIT Press, Cambridge, Mass.
Walker, K. 1994, 'Research with/in nursing: 'troubling' the field', Contemporary
Nurse, vol. 3, no. 4, p. 12.
Webb, D. n.d., Thinking About Suicide: Contemplating and comprehending the
urge to die, weblog, viewed 19 April 2015,
<http://thinkingaboutsuicide.org/crisis-of-the-self/>.
Wengraf, T. 2001, Qualitative Research Inteviewing: Biographic Narrative and
Semi-Structured Methods, Sage Publications, London.
White, M. 2007, Maps of narrative practice, W.W.Norton & Company, New
York.
White, M. & Epston, D. 1990, Narrative Means to Therapeutic Ends, W.W.
Norton & Company, New York.
Wing, J.K., Beevor, A.S., Curtis, R.H., Park, S.B., Hadden, S. & Burns, A. 1998,
'Health of the Nation Outcome Scales (HoNOS). Research and
development', British Journal of Psychiatry, vol. 172, no. 1, pp. 11-8.
Zola, I.K. 1993, 'Self, identity and the naming question: reflections on the
language of disability', Social Science and Medicine, vol. 36, no. 2, pp.
167-73.
References
191